SETTLED VERSUS RIGHT:
A THEORY OF PRECEDENT
Randy J. Kozel
Note to My Northwestern Readers. Thanks so much for your
willingness to share your reactions to these chapters, which are drawn
from my forthcoming book on precedent and stare decisis. I’ve included
the Introduction to give some background about the project, as well as
portions of Chapters 5 and 6, which in many respects are the core of
the book.
The book is just entering the publication process, which means I
have plenty of time to incorporate your suggestions. Thanks in advance
for your help!
CONTENTS
Introduction
1. Framing the Study of Precedent
2. The Stakes of Deference
3. Strength of Constraint
4. Scope of Applicability
5. Precedent and Pluralism
6. Precedential Strength in Doctrinal Perspective
7. Precedential Strength in Structural Perspective
8. Compromise, Common Ground, and Precedential Scope
9. Implications and Transitions
Conclusion
2
INTRODUCTION
The summer of 2005 was a time of transition at the U.S. Supreme
Court. Soon after the Court decided the last of its pending cases in
June, Sandra Day O’Connor announced her plan to retire. Justice
O’Connor had served on the Court since 1981. She made history as the
first female justice, and her trademark pragmatism made a deep
impression on American law. Around the halls of the Court, it is
commonly said that any time a justice departs, the institution is made
anew. With a jurist of Sandra Day O’Connor’s stature on the cusp of
leaving, that sentiment seemed as true as it had ever been.
Just two months after Justice O’Connor’s announcement came the
news that William Rehnquist had died. He had served on the Court
since 1972, taking over as Chief Justice in 1986. His legacy extended
beyond his legal decisions and into countless aspects of the Court’s
procedures and practices. Viewed alongside the retirement of Justice
O’Connor, Chief Justice Rehnquist’s passing foretold the end of one era
and the dawn of another. The most recent departure from the Court
had been that of Harry Blackmun, who retired in 1994. Now eleven
years later, the Court faced the loss of two justices of enormous
influence—who between them had served for more than fifty years—in
the course of only a few months.
Initially, Justice O’Connor’s seat on the Court was to be filled by
John Roberts, who was serving as a federal appellate judge. But after
Chief Justice Rehnquist’s death, President George W. Bush revised
Judge Roberts’s nomination. Judge Roberts would now take over as
Chief Justice, with Justice O’Connor’s successor—eventually, Samuel
Alito—to be selected later.
Supreme Court justices earn their appointments based on their
individual qualities and achievements. Upon their confirmation,
however, they join a tribunal with two centuries’ worth of practices,
customs, and decisions. A key issue for every new justice is how to
balance respect for the Court’s past with solicitude for its future.
That issue would arise in illuminating fashion during Judge
Roberts’s confirmation hearing before the Senate Judiciary Committee.
Like previous nominees, Judge Roberts was asked about the degree of
3
respect that is owed to the Supreme Court’s prior opinions—in other
words, its precedents. In American legal culture, courts commonly
describe precedents as carrying great weight. By respecting their
precedents, courts validate a time-honored principle: stare decisis et
non quieta movere, a Latin phrase meaning “[t]o stand by things
decided, and not to disturb settled points.” 1 The phrase, which is
commonly abbreviated as stare decisis, captures the idea that today’s
judges should not lightly disrupt the decisions of their predecessors.
Even so, it is always possible for a court to overrule its precedents, so
long as there is sufficient justification for doing so. The goal is to
preserve the law’s stable core without permanently entrenching every
judicial mistake.
During an exchange with Judge Roberts, Senator Arlen Specter
raised the topic of stare decisis in the context of Roe v. Wade (1973),
the Supreme Court’s landmark ruling on abortion rights. Yet the
Senator’s question went beyond Roe and addressed “principles of stare
decisis” more generally. Judge Roberts responded in kind. He began
with an appeal to history, explaining that America’s founders
“appreciated the role of precedent in promoting evenhandedness,
predictability, stability, [and] the appearance of integrity in the
judicial process.” He then turned to the Supreme Court’s modern
approach to precedent, which considers factors like whether prior
decisions have “proven to be unworkable” or “been eroded by
subsequent developments.”
Judge Roberts noted that to overrule a precedent is to give “a jolt
to the legal system.” At the same time, he cautioned that deference to
precedent is only presumptive, not absolute. It is true that the
overruling of precedent can tax the system. But sometimes “that’s a
price that has to be paid.” He illustrated with the example of Brown v.
Board of Education (1954), in which the Court broke from its past to
make clear that racial segregation in public services violates
fundamental constitutional precepts. 2
The experience of Judge Roberts—now Chief Justice Roberts—was
far from unique. Since his appointment, three more justices have taken
seats on the Court: first Samuel Alito, then Sonia Sotomayor, and
finally Elena Kagan. The role of precedent arose during each of their
1
2
BLACK’S LAW DICTIONARY (8th ed. 2004).
109th Cong. 141–44 (2005).
4
confirmation hearings. All three of them offered explanations similar
to that of then-Judge Roberts: the Court’s precedents warrant
meaningful deference, but such deference is not absolute. 3 And they
have continued to endorse this understanding of precedent upon taking
their positions on the Court. Indeed, every sitting justice has
acknowledged the importance of deferring to precedent under certain
circumstances. Each justice has also noted that precedent must
sometimes yield. The question is when.
That question has been at the center of many of the Court’s most
controversial rulings. It was there when the Court upheld the central
holding of Roe v. Wade. It was there when the Court rejected a
challenge to the Miranda warnings that police officers must give to
suspected criminals. More recently, it was there when the Court ruled
that the First Amendment affords strong protection to political ads by
corporations and labor unions—a ruling President Barack Obama
criticized during the 2010 State of the Union address not simply for
being wrong, but for having “reversed a century of law.” These disputes
over precedent are pervasive and important. They are also deeply
complex. The complexity reaches all the way down to the foundational
issue of why a judge would ever willingly accept a ruling she believes
to be wrong.
*
*
*
The study of precedent is the study of mistakes. Some past
decisions were misguided from the outset. Others began sensibly
enough but became shaky over time as facts changed. The issue in
either case is what to do next. Should today’s judges stand by prior
decisions they view as incorrect? Or should they set the record straight
and improve the law going forward?
At first glance the answer may seem obvious: judges should never
consciously repeat the mistakes of the past. But the calculus turns out
to be complicated. People might have made investments and modified
their behaviors in reliance on past judicial decisions. There is also the
worry that if judicial decisions are reversed too readily, the law will
lose its durability and impersonality and be reduced to whatever
111th Cong. 300 (2010) (J. Kagan); 111th Cong. 96–97 (2009) (J. Sotomayor); 109th
Cong. 318–19 (2006) (J. Alito).
3
5
today’s judges say it is. And it is always possible that, notwithstanding
the contrary belief of today’s judges, the previous decision actually
represents the more accurate interpretation of the law. In light of
possibilities like these, maybe it is better—at least sometimes—to let
things be.
As then-Judge Roberts noted during his confirmation hearing, the
Supreme Court has articulated a host of considerations to inform the
choice between retaining and jettisoning a decision that is incorrect in
the eyes of today’s justices. Key factors include the precedent’s
procedural workability, the soundness of its factual premises, the
extent to which subsequent decisions have eroded its foundations, and
the reliance it has generated. Still, the justices continue to disagree
over the role of precedent in particular cases. To some, the best
explanation for this disagreement is that stare decisis is really no
principle at all. On that account, fidelity to precedent seldom (if ever)
sways a justice from her preferred course. There is so much play in the
joints that even as they talk about stare decisis, the justices manage to
preserve the precedents they like and overrule the ones they don’t.
These sentiments occasionally come from the justices themselves.
Justice Scalia once criticized a majority opinion for treating the
doctrine of stare decisis as a “result-oriented expedient” rather than a
consistent principle. 4 A decade earlier, Justice Marshall directed a
comparable criticism at a majority opinion that upset settled law. He
concluded that “[n]either the law nor the facts” had changed; “[o]nly
the personnel of this Court did.” To Justice Marshall, the lesson was
clear: “Power, not reason, is the new currency of this Court’s
decisionmaking.” 5
Comments like these reflect a tension in the Supreme Court’s
treatment of precedent. While there is widespread agreement among
the justices about the factors that are potentially relevant to a dubious
precedent’s retention or overruling, there has been far less discussion
of how stare decisis fits into various theories of judging. Nor has the
Court devoted much attention to explaining why certain outcomes are
so problematic as to trigger prompt overruling, while others should be
tolerated in pursuit of values such as stability, continuity, and the
protection of settled expectations. The lack of a comprehensive
4
5
Lawrence v. Texas, 539 U.S. 558, 591 (2003) (Scalia, J., dissenting).
Payne v. Tennessee, 501 U.S. 808, 844 (1991) (Marshall, J., dissenting).
6
explanation can sometimes make it seem like the Court is being
inconsistent in its treatment of precedent. The effect is especially
pronounced within the realm of constitutional law, which draws the
Court into debates over the protection of fundamental liberties and the
essential structure of government. Some thirty years ago, Henry
Monaghan described the problem in terms that remain resonant today:
“Because a coherent rationale for the intermittent invocation of stare
decisis has not been forthcoming, the impression is created that the
doctrine is invoked only as a mask hiding other considerations. As a
result, stare decisis seemingly operates with the randomness of a
lighting bolt: on occasion it may strike, but when and where can be
known only after the fact. A satisfactory theory of constitutional
adjudication requires more than that.” 6
Without a meaningful role for precedent, the law sacrifices a large
share of its continuity, constraint, and impersonality. Decisions of the
Supreme Court become the products of fluctuating assemblages of
justices who come and go from the bench rather than the outputs of an
enduring institution that maintains its identity over time. The danger
is not that the overruling of precedent will lead to rioting in the streets
or widespread resistance to the Supreme Court’s edicts. The costs are
more in the nature of untapped potential. Time and again, the justices
have underscored that deference to precedent promotes the rule of law.
But those affirmations occur at the level of abstract theory. By
translating them into practice, the Court can bolster the idea that its
decisions flow from enduring legal principles rather than individual
proclivities, and that the Constitution truly is more than “what five
Justices say it is.” 7
Allow me to illustrate by reference to Citizens United v. Federal
Election Commission (2010), which I mentioned above and about which
I will have more to say in the pages ahead. In Citizens United, a fivejustice majority voted to overrule precedent by enlarging the First
Amendment liberties of corporations and labor unions. Four justices
resisted that result, but they fell one vote short. For now, let us reserve
judgment on whether the better argument was that of the five-justice
Henry Paul Monaghan, Stare Decisis and Constitutional Adjudication, 88 COLUM. L.
REV. 723, 743 (1988).
7 Lewis F. Powell, Jr., Stare Decisis and Judicial Restraint, 47 WASH. & LEE L. REV.
281, 288 (1990).
6
7
majority or the four-justice dissent. Instead, think about the impact of
the case going forward. Absent some presumption of deference to
precedent, whether Citizens United remains the law of the land—
which is to say, what the First Amendment means as applied to an
important area of campaign finance regulation—depends on whether
personnel changes at the Court turn the four-justice dissent into a fivejustice majority. Nor does the cycle end there. Assume that Citizens
United is reversed after a new justice arrives at the Court, but that in
short order a member of the majority coalition retires and is replaced
by a differently-minded justice. Without a meaningful doctrine of stare
decisis, the pendulum could just as easily swing back. All of this
despite the fact that the Constitution itself will not have changed a bit.
Citizens United suggests a broader point about the Supreme
Court’s role in the constitutional order. In 2016, Lawrence Norden
wrote in The Atlantic that “it is no exaggeration to say that the next
appointments to the Supreme Court will have a profound impact on
political power in the United States.” 8 The underlying premise is clear:
In modern constitutional law, the salient mechanism of change is not
the formal amendment process, but rather the appointment of new
justices to the Supreme Court.
This reality, I submit, is dispiriting and detrimental.
Constitutional principles should be overarching and enduring.
Deference to precedent advances the valuable ideal that it takes
something more than a group of nine (or, in a split decision, five)
individuals to declare what the Constitution requires. To be sure, the
identity and interpretive predilections of individual judges will always
matter. The composition of the courts will and should remain a topic of
interest to political campaigns and social movements. But the fact that
judges matter does not resolve the issue of how much they should
matter. A meaningful doctrine of precedent asks the individual judge
to subordinate—not always, but sometimes—her personal view of a
case to the historical practice of her court as an institution. Judges still
matter under a regime of stare decisis. They just matter less. And that
is a valuable thing in a system that aspires to promote the rule of law
as opposed to the rule of individual men and women. 9
Lawrence Norden, The U.S. Supreme Court Can Still Take Big Money Out of
Politics, ATLANTIC, Jan. 13, 2016.
9 Cf. Monaghan, Stare Decisis and Constitutional Adjudication, supra note __, at 752
(“A general judicial adherence to constitutional precedent supports a consensus about
8
8
This book develops a theory of precedent designed to enhance the
stability and impersonality of constitutional law. The problem with the
Supreme Court’s current approach to precedent is not that the justices
are behaving in an unprincipled manner. The problem is that the
modern doctrine of stare decisis is undermined by principled
disagreements among justices acting in good faith. The doctrine’s
structure and composition all but guarantee that conclusions about the
durability of precedent will track the justices’ individual views about
whether decisions are right or wrong and whether mistakes are
harmful or benign. To rehabilitate the doctrine of stare decisis so it can
bridge philosophical divides, we need to rethink the way in which
precedent interacts with constitutional theory.
The starting point is recognizing the implications of a basic fact
about our legal culture. We have not reached anything approaching
consensus regarding the proper method for understanding and
applying the Constitution. Rather, ours is a second-best world of
pervasive disagreement over constitutional interpretation. That
requires a second-best theory of stare decisis attuned to the challenges
of judicial disagreement and the value of precedent in overcoming
them. It remains possible for stare decisis to play the vital role the
Supreme Court has described for it in enhancing the continuity and
impersonality of constitutional law. For that to occur, we need to
reconsider the doctrine from the ground up. The prevailing approach to
precedent implies a greater degree of agreement about constitutional
theory than actually exists. If stare decisis is to fulfill its promise, we
must account for the unique challenges posed by disagreements—goodfaith, principled disagreements—about the proper ends and means of
constitutional interpretation.
*
*
*
Having foreshadowed the arguments toward which the book will
build, allow me to circle back to explain the path it will take.
In the first part of the book, I aim to provide a descriptive and
analytical account of precedent that is independent of the normative
claims that will come later. Chapter 1 begins by distinguishing two
the rule of law, specifically the belief that all organs of government, including the
Court, are bound by the law.”).
9
common situations: those in which a court is considering the effect of
its own prior decisions, and those in which a court must apply
decisions from a tribunal of superior rank. The former scenario
involves what are sometimes called horizontal precedents; the latter
involves precedents that operate vertically, running from higher courts
to lower courts. Though both situations deal with the impact of prior
decisions on later courts, they are governed by different rules in the
U.S. federal system. While a court always has the power to reconsider
its own past decisions, lower courts do not enjoy comparable discretion
to revise the opinions of higher courts. The Supreme Court has insisted
on this point, making clear that lower courts may never reject a
Supreme Court decision—even if the decision is obviously flawed, has
been eroded over time, or has been called into question by the justices
themselves. These different rules require distinguishing vertical and
horizontal precedents even while recognizing that some of the
arguments for (and against) deference will overlap.
After drawing a line between vertical and horizontal precedents,
Chapter 1 turns to another pivotal distinction, this one between
precedential strength and precedential scope. In evaluating the role of
precedent, it can be tempting to focus exclusively on the degree of
constraint that prior decisions exert on future disputes. It is a
precedent’s strength that ultimately determines whether there is a
sufficient justification for overruling it. Yet strength is only part of the
story. No matter how strong a precedent is deemed to be, the precedent
has no constraining force in situations it does not reach. There must be
a threshold determination whether a prior decision applies to a later
case. Sometimes it is quite clear that one case governs another, so the
only valid options for the later court are to reaffirm or overrule. But in
many other cases, whether a precedent applies to the case at hand is a
thorny and contentious question. Keeping in mind these dual
considerations of strength and scope is crucial to analyzing and,
hopefully, improving the treatment of precedent.
Next, I introduce two more sets of distinctions that are helpful in
understanding the law of precedent. The first is the type of case a court
is called upon to resolve. Conventional wisdom holds that judicial
interpretations of statutes are entitled to maximum deference going
forward, whereas interpretations of the Constitution receive weaker
deference. I offer some reasons for being skeptical about this
distinction, and I argue that in all events, the fact that constitutional
10
precedents receive relatively weak deference under existing law does
not mean such deference is weak in absolute terms. Even if statutory
cases receive the most insulation from overruling, that leaves a broad
range of possibilities for how much deference should attach to
constitutional decisions. The intricacies of constitutional stare decisis
will be my focus for much of the book, though many aspects of my
analysis will apply to statutory (and common-law decisions) as well.
The remainder of Chapter 1 surveys the various functions that
precedents serve in modern American law. Precedents are means of
transmitting knowledge from past to present, so they can improve
judicial decisionmaking even when there is no obligation to follow
them. In some cases, though, it is not left to the later court to make up
its mind about whether to follow precedent. Instead, the later court is
duty-bound to stand by the decision of the earlier court. This is easiest
to see in the context of vertical precedent, as when a federal trial court
is required to follow a Supreme Court decision despite reservations
about that decision on the merits. Precedent can also constrain future
iterations of the court that issued it. The Supreme Court is properly
understood as constrained to follow its precedents under certain
conditions: namely, when the Court’s articulated criteria for overruling
are not satisfied. This constraining function presents both the
strengths and weaknesses of precedent-based judging in their starkest
form. At its best, precedent limits the discretion of subsequent judges
and contributes to a stable, consistent, and impersonal system of law.
Yet a strong doctrine of precedent can also lead to the repetition and
entrenchment of earlier judges’ miscues. These are the stakes of the
debate.
I elaborate on the stakes in Chapter 2, which begins by chronicling
some of the commonly-cited benefits of deference to precedent. They
include the conservation of judicial resources, the protection of settled
expectations, and the preservation of a stable environment to facilitate
planning. They also include impersonality. A commitment to precedent
can encourage the equal treatment of litigants, reducing the extent to
which the idiosyncrasies of their situations affect the outcome of their
disputes. At the same time, deference to precedent can allow the law to
transcend the identity of the judge who happens to be presiding over a
particular case. If a judge must follow precedent, her individual
preferences and tendencies become less salient.
11
On the other side of the scale are the costs of abiding by precedent.
Imagine that five justices of the Supreme Court conclude a prior
decision reflects an erroneous understanding of the Constitution.
Those justices also happen to be stalwart proponents of stare decisis,
for reasons including continuity and impersonality. They accordingly
vote to reaffirm the decision notwithstanding their misgivings about
its rationale. While they believe themselves in possession of a sound
basis for doing so, the justices relinquish the opportunity to replace
(what they believe to be) an incorrect rule with a more accurate one.
They consciously allow a mistake to go uncorrected. I will end up
defending a meaningful doctrine of precedent notwithstanding these
countervailing considerations. But the costs must be appreciated if the
doctrine of stare decisis is to strike the appropriate balance between
continuity and change.
Before closing the second chapter, I offer a few words about the
consistency of stare decisis with the Constitution. Issues of legitimacy
are complicated and fascinating, but I do not dwell on them for the
simple reason that they are uncontroversial in modern judicial practice.
Justices of the Supreme Court vary in their readiness to overrule
flawed decisions, but no justice has challenged the lawfulness of stare
decisis. Still, a few commentators have raised such a challenge, so I
briefly examine some possibilities for defending the legitimacy of stare
decisis in constitutional cases. Those possibilities draw on the
Constitution’s text, the background understandings and practices in
place at the time of the founding, the structure of the federal judiciary,
and the need for judges to act in a collective, cooperative fashion
notwithstanding their interpretive disagreements.
Chapters 3 and 4 unpack the complementary concepts of
precedential strength and precedential scope. I explain how both
concepts operate under the Supreme Court’s existing approach to
precedent, and I emphasize how they are shaped by underlying
conclusions about the ends and means of constitutional interpretation.
To begin with precedential strength: Nearly a century ago, Justice
Louis Brandeis described the tension inherent in the doctrine of stare
decisis as pitting the importance of leaving the law settled against the
value of getting the law right. 10 This characterization has endured, and
Burnet v. Coronado Oil & Gas Co., 285 U.S. 393, 406 (1932) (Brandeis, J.,
dissenting).
10
12
for good reason. In deciding whether to overrule a flawed decision, it is
natural to inquire into the bad effects the decision has created and to
predict the beneficial effects that would accompany a change of
direction. But the factors that make a decision good or bad are neither
static nor universal. They depend on the interpretive theory that a
particular judge adopts. For some judges, a prior decision’s
implications for justice and fairness shape whether it is harmful or
benign. Other judges treat those considerations as legally irrelevant.
Likewise, some judges measure the severity of a mistaken
interpretation based on how sharply it departs from the Constitution’s
original meaning at the time of the founding. Others find the
Constitution’s original meaning to be less relevant than considerations
such as a decision’s pragmatic or moral ramifications.
The point is simply that judges rely, whether explicitly or
implicitly, on their theories of interpretation to determine whether a
prior decision is correct in its reading of the Constitution. This can and
does lead to principled disagreements. If some Supreme Court justices
focus on the Constitution’s original meaning while others focus on
contemporary mores or policy judgments, it should be unsurprising if
they part ways over the soundness of certain precedents. Those same
variances in interpretive philosophy also inform the subsequent—and
distinct—determination of precedential strength, which dictates
whether a prior decision should be reaffirmed despite its flaws. Every
judicial decision has a host of consequences, ranging from on-theground practical effects to broader implications for governmental
design and political morality. Determining what types of consequences
are legally relevant depends on a given judge’s interpretive philosophy.
In turn, assessing whether a prior decision is so problematic as to
warrant overruling requires analyzing the decision’s legally relevant
implications while excluding other matters. That enterprise is
necessarily shaped by interpretive philosophy, and it will look different
depending on a judge’s methodological and normative commitments.
The same is true of a precedent’s scope of applicability, which I
discuss in Chapter 4. Evaluating whether a prior decision is relevant
to a newly arising dispute requires determining what the prior decision
means. In making that determination, a common step is to draw a line
between judicial statements that were necessary to a case’s resolution
and statements that were dispensable, with the former representing
the decision’s holding and the latter mere dicta. That distinction
13
informs the definition of precedential scope: Holdings are entitled to
deference in future cases, whereas dicta are nonbinding and may be
accepted or rejected at the pleasure of the subsequent court.
Notwithstanding its historical pedigree, the holding/dicta
distinction fails to explain existing federal practice, including at the
Supreme Court. While the Court occasionally insists on a strict line
between binding holdings and dispensable dicta, it regularly defers to
aspects of its opinions—including sweeping rules and doctrinal
frameworks—that range beyond the application of specific law to
concrete fact. Whether this phenomenon should be lauded or jeered
depends on underlying beliefs about the judicial role, the requirements
of the Constitution, and the utility of precedent in constraining
subsequent decisionmakers. Some interpretive philosophies seek to
minimize the extent to which judicial pronouncements displace
considerations such as the original meaning of the Constitution’s text.
On those theories, it is sensible to construe precedents narrowly. Other
theories make greater use of precedent as a tool of judicial constraint
or a source of common ground among differently-minded judges,
supporting the view that precedents should be defined in relatively
broad terms. These are only two of several possibilities I will discuss,
but they introduce the broader point. Just as attitudes toward the
strength of precedent are bound up with underlying interpretive
preferences, so too are accounts of precedential scope.
With the relationship between precedent and constitutional
philosophy established, the book moves from why it is important to
reconsider the role of precedent to how that reconsideration should
proceed. In pursuing this inquiry, my focus is the operation of stare
decisis at the U.S. Supreme Court. Over the past three decades the
justices have devoted considerable effort to discussing why precedent
deserves presumptive respect and why that presumption must
sometimes yield. In Chapters 5 and 6, I address the various factors the
Court has enumerated to guide its applications of stare decisis. Of
particular interest is the extent to which those factors possess objective
content that does not depend on an individual judge’s interpretive
philosophy. Separating stare decisis from disputes over constitutional
interpretation is vital if judicial responses to precedent are to
transcend individual beliefs about how best to understand the
Constitution’s teachings.
14
Stare decisis can bolster the stability and impersonality of
constitutional law only if it sometimes requires a justice to accept an
outcome she thinks is incorrect. If the pull of precedent gives way
every time a justice concludes a prior decision is wrong, the impact of
stare decisis dissipates. Yet that is precisely what we should expect
from a doctrine of stare decisis that allows the interpretive
philosophies of individual justices to dictate whether a prior decision is
reaffirmed or overruled. As I explain in Chapter 5, because a justice’s
interpretive philosophy colors her determination of which
considerations are legally relevant, it also goes a long way toward
informing her applications of stare decisis under existing law.
This concern would be less pressing if there were widespread
agreement about the proper ends and means of constitutional
interpretation. Imagine a Supreme Court comprised of nine justices
who agree about how the Constitution ought to be interpreted,
including which types of considerations are legally relevant and which
are not. Imagine that the justices also agree about what makes a
flawed precedent particularly bad—perhaps, for instance, that it
creates serious injustice. That consensus would open the door for a
consistent and systematic approach to precedent. Plessy v. Ferguson
(1896), which validated racial segregation in public accommodations,
would furnish a ready example of a flawed decision that was too unjust
to tolerate. 11 By contrast, a case like National Bellas Hess v.
Department of Revenue (1967), which arguably misconstrued the
authority of states to impose tax obligations on out-of-state sellers,
might be retained; even if the decision is incorrect, it is difficult to
construe it is immoral. 12 More generally, the justices’ conclusions about
the durability of precedent would continue to depend on their theories
of constitutional interpretation. But because those theories would be
universally held—at the Court, at least—they would facilitate a
consistent approach to precedent.
Now relax the assumption that the justices are in harmony, and
assume instead that they are sharply divided over the appropriate
methods of constitutional interpretation. The most obvious effect of
disagreement is that the justices will split over whether certain
precedents are wrong or right. But they will also disagree about
another point: the factors that make a flawed precedent not simply
11
12
163 U.S. 537.
386 U.S. 753.
15
wrong, but in need of overruling. That latter debate will tend to track
the justices’ differences of opinion over the appropriate principles for
interpreting the Constitution. In other words, there will be two points
of fracture. One relates to the characterization of a precedent as
incorrect, and the other relates to the considerations that justify a
flawed precedent’s overruling. Yet both inquiries will be informed by
the same methodological and normative priorities that divide the
justices in the first place. Whether a precedent is overruled will depend,
at base, on the interpretive philosophy that commands a majority of
sitting justices.
This is not how stare decisis is supposed to work. The reason why
the Supreme Court often links precedent to the rule of law is because
deference to past decisions can unite justices of varying interpretive
stripes. Two (or three, or nine) justices may disagree about how the
Constitution should be interpreted but still share a common dedication
to precedent. It draws together justices who would otherwise disagree
on the merits. But when the decision to overrule tracks the
interpretive preferences of individual justices, the connection between
stare decisis and judicial impersonality is severed. Precedent stops
serving as common ground for overcoming philosophical disagreements.
Invocations of stare decisis restate disagreements instead of bridging
them.
A comparable analysis applies to the definition of a precedent’s
scope of applicability. Defining a precedent’s contours depends in
significant part on interpretive preferences: how much one values
uniformity and guidance, the extent to which one is comfortable
displacing the best interpretation of the Constitution’s text, and so on.
Pervasive disagreements over constitutional theory create challenges
in fashioning a consistent account of precedential scope.
The key to developing better approaches to precedential strength
and precedential scope is acknowledging the impact of deep-seated
disagreements among judges about the proper way to interpret the
Constitution. In our world of pervasive interpretive disagreement, we
need to think about the role of precedent differently than we would
under conditions of widespread interpretive harmony. The question is
no longer which factors are potentially relevant to a precedent’s
retention or overruling. The inquiry must be narrowed to include only
those factors that are susceptible of principled application by justices
16
across the philosophical spectrum. We need a second-best theory of
stare decisis to complement our second-best world of interpretive
disagreement.
This position may seem counterintuitive, for it requires ignoring
certain considerations that would be relevant to a precedent’s
durability under conditions of interpretive agreement. Even so,
disregarding some of those considerations and cabining others is
necessary for stare decisis to overcome interpretive disputes rather
than repackaging them. The objective of this reconceptualization is
neither to increase nor decrease the power of precedent in any given
case. It is to disentangle a precedent’s correctness on the merits from
its claim to deference notwithstanding its flaws.
Chapter 6 applies these principles to determinations of
precedential strength. The approach I defend has some features in
common with the doctrine of stare decisis that currently prevails at the
Supreme Court. Several factors loom large on both accounts: a
decision’s procedural workability, the accuracy of its factual premises,
and the reliance it has yielded. Even so, using precedent to bridge
judicial disagreements means fine-tuning those factors to ensure that
their invocation does not collapse into disputes over interpretive
philosophy.
Though revising the inquiry into considerations such as
workability and factual accuracy is important, the most significant
change I propose relates to a precedent’s substantive effects. As I have
suggested, the relevance of such effects depends on one’s theory of
constitutional interpretation. Some theories prize matters of justice
and morality, others pragmatic results, still others compatibly with
founding-era understandings. If these are the drivers of whether a
precedent is overruled, the application of stare decisis will track
interpretive philosophies that differ from justice to justice. There is
nothing unprincipled about such a regime. Each justice might make
decisions about precedent that are consistent with her overall
interpretive philosophy. Still, this vision of stare decisis relinquishes
the ability to draw together justices who are sympathetic to different
interpretive schools. In so doing, it gives away a large share of the
promise of precedent.
17
Because their impact depends on contested matters of interpretive
philosophy, substantive effects must generally be excluded from the
stare decisis calculus. Allowing substantive effects to guide the
analysis all but guarantees that the treatment of precedent will be
bound up with deeper methodological disputes. Evaluating a
precedent’s substantive effects might be appropriate in a world of
interpretive agreement, in which the justices work from a shared set of
assumptions about which of a prior decision’s consequences are legally
relevant. But that is not our world. In our second-best world of
interpretive pluralism, a precedent’s substantive effects should bear on
its retention only in a small category of exceptional cases. The category
is comprised of decisions that an individual justice views as not simply
wrong or bad, but extraordinarily harmful. Of course, each justice
must make that assessment based on her individual interpretive
preferences. As a result, different justices will reach different
conclusions about which precedents fit the bill. Once a justice decides
that a mistaken precedent is responsible for causing extraordinary
harm, she is justified in refusing to stand by that precedent for the
sake of continuity. This exception coheres with the common
understanding of stare decisis as significant but nonabsolute.
Sometimes a precedent is too bad (from the perspective of an
individual justice) to tolerate. In those cases, the pull of precedent
gives way.
While the exception for extraordinary harm contemplates
occasional situations in which individual attitudes toward
constitutional interpretation are paramount, its narrowness reinforces
the importance of compromise in the ordinary course. A robust doctrine
of precedent regularly calls upon the justices to subordinate their
individual conclusions in deference to the Court’s institutional history.
Disregarding a precedent’s substantive effects goes hand-in-hand with
the idea that it takes more than disagreement to justify a precedent’s
overruling. Only when a decision strikes a justice as so inordinately
harmful that it cannot be tolerated does this principle yield.
The picture that emerges is one of precedent serving as a source of
common ground among differently-minded justices, one that facilitates
impersonal decisionmaking and coordinated action. The text of a
statute or constitutional provision provides a useful source of common
ground. There is no need for two justices to argue about whether, say,
each state should have the same number of Senators. The
18
Constitution’s clear text furnishes the answer. It will furnish the
answer when there are five justices on the Court who give primacy to
the Constitution’s original meaning. It will furnish the same answer
when there are five justices who view the Constitution as a living
document that evolves over time.
Precedent can play a similar role. It allows some points to be taken
as given rather than perpetually debated. And it is does so in a fashion
that is fundamentally neutral. To be sure, nearly every precedent has
its backers and its critics. But the general practice of precedentfollowing does not work exclusively to the advantage of living
constitutionalists, or originalists, or anyone else. Some precedents are
consistent with the Constitution’s original meaning. Others are
consistent with living constitutionalism. This is important, because it
limits what the individual justice sacrifices by committing herself to
precedent. It is not as if a justice is asked to jettison her own
interpretive philosophy and pledge fidelity to another. Rather, she
agrees to defer to the Court’s precedents, some of which she will favor
and some of which she will not.
I build upon these themes by considering more directly the
decisionmaking process of the individual jurist. Up until this point, I
defend second-best stare decisis on the assumption that the justices of
the Supreme Court are jointly committed to maintaining a meaningful
doctrine of precedent. I think this assumption is sound: Justices across
the methodological spectrum have emphasized the importance of stare
decisis to a stable and impersonal rule of law, and I see no reason to
doubt their sincerity or resolve. Even so, I relax this assumption and
contend that a justice is well served to cast her lot with stare decisis
irrespective of whether her judicial peers—present or future—follow
suit. By adhering to precedent, even a single justice can promote
fundamental values of continuity and impersonality while contributing
to the entrenchment of stare decisis as an ongoing practice.
All of this depends on the effectiveness of attempting to limit the
impact of factors such as a precedent’s substantive effects. Chapter 7
discusses an alternative approach to precedential strength grounded
not in the substance of stare decisis doctrine, but rather in the
structure of Supreme Court decisionmaking. The proposal is to require
a supermajority vote in order to overrule a precedent. The rationale is
straightforward. The more votes it takes to overrule, the more likely it
19
becomes that an overruling will require cooperation among justices
who have different theories of constitutional interpretation. And the
greater the likelihood that such cooperation is required, the lower the
chances that a precedent will be jettisoned due to nothing more than
personnel shifts—and accompanying changes in the Court’s
interpretive locus.
While a supermajority voting rule has the potential to enhance the
impersonality of judicial decisionmaking, I will suggest that it is less
promising than the doctrinal revisions to stare decisis as described in
Chapter 6. The supermajority rule operates by increasing the number
of individual perspectives that are necessary in order to change the law.
But it does not ask the individual justice to subordinate her own
interpretive preferences to the role of the Court as an institution. That
is a missed opportunity for reinforcing the prevalence of the rule of law
over the rule of individual women and men.
In Chapter 8, I turn to the implications of interpretive
disagreement for the definition of a precedent’s scope of applicability.
As a descriptive matter, debates over precedential scope cannot be
reduced to distilling the holding of a case and separating it from the
dicta. The Supreme Court commonly accords deference to aspects of its
opinions that range far beyond the narrow application of law to fact.
The legitimacy and desirability of that practice depend on underlying
beliefs about constitutional interpretation and judicial decisionmaking.
The problem is familiar by now. Those underlying beliefs vary greatly
from judge to judge and justice to justice, yet the doctrine of stare
decisis must be grounded in considerations that steer clear of
contestable interpretive and normative commitments.
In light of the challenges posed by interpretive pluralism, I urge a
revised approach to precedential scope that leverages areas of
agreement within existing law. With occasional exceptions, it is
generally accepted by Supreme Court justices across the
methodological spectrum that judicial asides and hypotheticals are not
entitled to deference in future cases. It likewise is generally accepted
that a judicial opinion can establish a precedent by setting forth a
doctrinal rule or framework even if the rule or framework obviously
ranges beyond the facts presented to the court for resolution. Given
their widespread acceptance, these two principles provide building
20
blocks for an approach to precedential scope that adds consistency
while respecting what has gone before.
A more difficult question involves a prior court’s statement of its
reasoning. The Supreme Court sometimes treats decisional rationales
as entitled to deference in future cases. At other times, the Court
defers only to a decision’s legal rule, characterizing expressed
rationales as extraneous. In the face of this divide, I urge a
compromise: The reasons offered in support of a decision do not
warrant deference in all cases, but neither may they be discounted in a
way that undermines the logic of the decision that announced them.
This distinction, imperfect as it might be, responds to interpretive
pluralism by walking the line between excessive deference to
peripheral statements and inadequate respect for the considered
expressions of prior courts.
Chapter 9 asks what the theory of second-best stare decisis means
for leading schools of constitutional interpretation. I discuss the ways
in which second-best stare decisis coheres with and complements
certain interpretive philosophies, as well as the ways in which it
challenges
them.
Leading
methodologies
such
as
living
constitutionalism and originalism are compatible with second-best
stare decisis in significant respects, including the use of precedent to
guide and constraint future decisionmakers. Yet tensions will certainly
arise. The living constitutionalist will sometimes be asked to stand by
an originalist decision that she views as inconsistent with
contemporary mores or sound policy judgments. The originalist will
sometimes be asked to validate a decision that she views as having
departed from the original meaning of the Constitution’s text.
Despite these costs, deference to precedent is justified by the ideal
of the Supreme Court as an enduring institution rather than the
contingent product of individual predilections. There is also a more
practical consideration at work: If living constitutionalists are unlikely
to convince many originalists to join their cause, and if originalists are
similarly unlikely to convince many living constitutionalists to come
aboard, perhaps the best approach is one that gives something to—and
asks something from—both.
The book concludes with brief parting thoughts about the
relationship between precedent, impersonality, and continuity in
21
modern constitutional law. The reality of American law and politics is
that constitutional change happens through judicial appointments, not
formal amendments. This is entirely reasonable. The people elect their
president, and each president tries to select Supreme Court justices
who view the law in a particular way. This does not imply the Court is
a “political” institution, at least if that word is taken to mean the
justices make their decisions based on political preferences. There is
room for principled disagreement in the interpretation of the
Constitution.
When some justices leave the Court and others arrive, the
dynamics can change such that interpretive approaches that formerly
were in the minority come to predominate. Again, this is unremarkable.
Nor should it surprise us if a shift in the Court’s interpretive locus
leads to the reconsideration of precedents that reflect now-disfavored
ways of understanding the Constitution. But ebb and flow is not the
only way to design a system of constitutional adjudication. As an
alternative, we can imagine a system in which constitutional law
retains a stable, continuous core even as individual justices come and
go. That is the world of second-best stare decisis.
Legal continuity comes at a cost that must be acknowledged. Stare
decisis means tolerating interpretations that one believes to be
mistaken. It also means declining opportunities to innovate when the
Court has already resolved an issue. Like the turtles adorning this
book’s cover—turtles featured on the lampposts that ring the Supreme
Court in Washington, D.C.—a court that commits itself to stare decisis
will proceed incrementally, deliberately, slowly in pushing the law
forward. It will continue to fashion new rules in cases of first
impression. But where the law is settled, it will tend to leave things as
they stand. Legal change occurs not through the courts, but through
other channels: channels like the enactment of state and federal
legislation to protect important rights, and the proposal of
constitutional amendments for national consideration and debate.
Where the political process cannot or will not act, the law generally
remains intact—even at the cost of enduring a past mistake.
The costs of continuity are real, but so are the benefits. The
potential vacillation of constitutional law following changes in judicial
personnel is replaced by an abiding sense of stability and
impersonality. By deferring to precedent, the justices subordinate their
22
individual perspectives to the Court’s institutional identity. They
establish their commitment to an enduring institution that cannot be
reduced to the tendencies of a sitting majority. And they make good on
the promise of the Constitution as more than what five justices say it
is. The judges change, but the law remains the same.
23
CHAPTER 5: PRECEDENT AND PLURALISM
The previous two Chapters examined how constitutional philosophy
affects approaches to precedent. A Supreme Court justice who adopts a
general practice of deferring to precedent faces two principal questions
of implementation. One is how strongly to defer. The other is how to
determine when a precedent applies and when it does not. The
answers to these questions depend on underlying matters of
interpretive theory, normative commitment, and constitutional
understanding.
Determining which precedents are vulnerable to overruling
requires explaining why it is important for the law to be correct in the
first place. This turns out to be complex and controversial. For some,
accurate interpretation is valuable because it promotes popular
sovereignty as exercised through the people’s ratification of the
Constitution. For others, key considerations include morality and
justice as understood in light of contemporary mores. Still others focus
on consequentialist benefits and effective social policy. And the list
goes on. To figure out which decisions should be stricken from the
books even at the expense of continuity, we need to know both how to
identify judicial mistakes and how to figure out which mistakes are
worse than others.
These debates are more than theoretical; they also implicate the
treatment of precedent in practice. When a justice determines that the
Court went astray in recognizing the constitutional right of
corporations to advocate for political candidates, how should she assess
the harm that would result from leaving the offending precedent on the
books? Is it a matter of consequentialist analysis? Does it depend on
the precedent’s effects on popular sovereignty? Do moral judgments
have some role to play? Questions like these are crucial to determining
the magnitude of a constitutional mistake, yet they admit of no answer
until they are connected with a deeper interpretive theory. 13
If a justice’s interpretive methodology emphasizes to factors such as
freedom from governmental oppression, a case like Plessy v. Ferguson
(1896) will be in urgent need of overruling given its endorsement of
See Kurt T. Lash, The Cost of Judicial Error: Stare Decisis and the Role of
Normative Theory, 89 NOTRE DAME L. REV. 2189, 2205 (2014).
13
24
racial segregation. 14 But a case like Miranda v. Arizona (1966), even if
deemed incorrect in its approach to police questioning, is much more
difficult to view as oppressive to individuals. 15 Any argument for
overruling would need to invoke other considerations. Alternatively, if
a justice’s interpretive methodology is grounded in popular sovereignty,
the denial of a constitutional right “to nondiscriminatory treatment by
a private employer” might be tolerable, because the political process
theoretically can provide legislative protections against employment
discrimination when the courts have failed to act. 16 Yet the same
denial could be problematic for theories centered on individual fairness
or equal protection.
These types of questions abound in the application of stare decisis.
To take just a few examples:
•
If a justice believes the Supreme Court incorrectly withheld
constitutional protection from same-sex couples for their
private relationships, what metric should she use to evaluate
the harmfulness of leaving the erroneous precedent intact? 17
•
If a justice believes the states possess broad powers to impose
tax-collection obligations on out-of-state sellers, how should
she evaluate the harm caused by a precedent that unduly
limits those powers? 18
•
If a justice believes Roe v. Wade (1973) was mistaken in
recognizing a constitutional right to abortion, how should she
weigh the ramifications of retaining Roe versus overruling
it? 19
To answer these questions and others like them, we need an
organizing theory to tell us which effects of precedent are legally
salient. That was the thesis of Chapter 3. A justice’s interpretive
163 U.S. 537.
384 U.S. 436.
16 Kurt T. Lash, Originalism, Popular Sovereignty, and Reverse Stare Decisis, 93 VA.
L. REV. 1437, 1459 (2007).
17 See Lawrence v. Texas, 539 U.S. 558 (2003).
18 See Quill Corp. v. North Dakota, 504 U.S. 298 (1992).
19 See Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833
(1992).
14
15
25
theory is what determines whether the effects of a flawed decision are
legitimate reasons for overruling it, or rather consequences that are
lamentable but ultimately inapposite to the judicial process.
The same goes for a precedent’s scope of applicability, as explained
in Chapter 4. A justice who thinks precedents warrant presumptive
deference needs to figure out when a prior decision is relevant to the
case at hand. A potential starting point is the distinction between
binding holdings and dispensable dicta. But the question remains why
that distinction is appropriate for defining the scope of precedent—a
question whose difficulty is exacerbated by the lack of a consistent
approach at the Supreme Court. Any answer will imply a particular set
of understandings about the manner in which the Constitution ought
to be interpreted and implemented.
All of this would be complicated enough if the justices were in
perpetual agreement about the precepts of constitutional
interpretation. In such a scenario, each justice would still occasionally
disagree with her peers, past and present alike. For example, two
originalist justices might reach divergent conclusions about the
original meaning of a particular provision of the Constitution. Or two
pragmatic justices might disagree about whether a constitutional rule
is effective in practical terms. Notwithstanding these divergences, the
justices would be in basic agreement about what it means to interpret
the Constitution faithfully and competently.
Overlapping interpretive and normative commitments would also
facilitate a unified approach to precedent. The justices might rally
around the view that a flawed precedent is most troubling when it
offends popular sovereignty, or when it violates contemporary mores,
or otherwise. Similarly, they might agree that the need for constraint
justifies defining precedents in broad terms, or that a proper
understanding of the judicial role requires limiting precedents to their
essential applications of narrow rules to concrete facts. Whatever the
content of their interpretive agreements, the justices would be poised
to pursue a consistent vision of precedent.
Of Harmony and Discord
In reality, American constitutional practice is awash in interpretive
disagreements. This is true of the federal judiciary as a whole, and it is
26
true of the Supreme Court in particular. Justice Scalia noted in 2013
that the justices were not “in agreement on the basic question of what
we think we’re doing when we interpret the Constitution.” 20 The extent
of disagreement becomes even more pronounced when we consider
shifts in interpretive philosophy that occur over the years as individual
justices come and go.
The modern Supreme Court does not consistently adopt any
particular methodology of constitutional decisionmaking. Take, for
instance, the originalist school of interpretation. Some justices have
been clear about their adherence to originalism, and the Court often
refers to the Constitution’s original meaning in the course of
explaining its decisions. Occasionally, originalism even takes center
stage. A prominent example is the Court’s recent ruling that the
Second Amendment protects an individual’s right to possess firearms.
There, a majority of justices joined an opinion adopting “the original
understanding of the Second Amendment.” 21
Notwithstanding this endorsement of originalism, in other cases
the Court resolves constitutional questions with little or no attention
to original meanings. Sometimes the Court’s analysis is steeped in
political theory, such as the belief that an unfettered marketplace of
ideas is the lynchpin of expressive liberty. The field of campaignfinance law provides a ready illustration. When the Supreme Court
ruled in Citizens United v. FEC (2010) that corporations cannot be
barred from candidate advocacy, the majority opinion was driven by
conceptual arguments about what the First Amendment “stands
against”: namely, “attempts to disfavor certain subjects or
viewpoints.” 22 The majority also invoked the Constitution’s original
meaning in concluding “[t]here is simply no support for the view that
the First Amendment, as originally understood, would permit the
suppression of political speech by media corporations.” 23 But the crux
of the decision was the determination—not obviously linked to any
investigation of original meanings—that “independent expenditures,
including those made by corporations, do not give rise to corruption or
Jennifer Senior, In Conversation: Antonin Scalia, N.Y. MAG. (Oct. 6, 2013),
http://nymag.com/news/features/antonin-scalia-2013-10/.
21 District of Columbia v. Heller, 554 U.S. 570, 625 (2008).
22 558 U.S. 310, 340.
23 Id. at 353.
20
27
the appearance of corruption.” 24 It was left to a concurrence to conduct
a deeper inquiry into the decision’s consistency with the original
meaning of the First Amendment. 25
In other instances, the justices base their decisions on assessments
of perceived constitutional purposes or longstanding political
traditions. An example arose in 2014 when the Court interpreted the
Recess Appointments Clause. A majority of justices noted that “in
interpreting the Clause, we put significant weight upon historical
practice” and “must hesitate to upset the compromises and working
arrangements that the elected branches of Government themselves
have reached.” 26
There are also cases in which contemporary moral sensibilities are
prevalent, as exemplified by the Court’s recognition of a constitutional
right to same-sex marriage in Obergefell v. Hodges (2015). The
majority in Obergefell made clear its analysis was informed by evolving
constitutional norms. It noted that “[t]he history of marriage is one of
both continuity and change,” and it added that “[t]he nature of
injustice is that we may not always see it in our own times.” 27 While
the Court’s prior “cases describing the right to marry presumed a
relationship involving opposite-sex partners,” they did not entrench
such a view for time immemorial. Instead, the Obergefell majority
revised the Court’s caselaw in order to pursue “a better informed
understanding of how constitutional imperatives define a liberty that
remains urgent in our own era.” 28
The point is that the Court regularly shifts between interpretive
approaches without suggesting its “different methods are reducible to
one master method,” much less furnishing a passkey for undertaking
such a decryption. 29 I will refer to this approach as pluralistic,
reflecting a vision of constitutional decisionmaking characterized by
the absence of commitment to any particular interpretive theory.
Id. at 357.
See id. at 385 (Scalia, J., concurring).
26 National Labor Relations Board v. Noel Canning, 134 S. Ct. 2550, 2559–60. (2014).
27 135 S. Ct. 2584, 2595, 2598.
28 Id. at 2602.
29 Stephen M. Griffin, Pluralism in Constitutional Interpretation, 72 TEX. L. REV.
1753, 1757 (1994).
24
25
28
To believe multiple styles of constitutional argument are legitimate
is not necessarily to be a pluralist. The Court might conclude that, for
example, some constitutional provisions are properly understood in
light of their original meanings, while others entail consulting
contemporary sensibilities. The Court would also need to explain the
underlying normative basis of a theory that points toward original
meanings in some cases and contemporary mores in others. No such
explanation has been forthcoming. For better or worse, the Court has
been willing to emphasize different argument styles from case to case
without presenting an overarching theory grounded in a defined set of
normative values. It has, in short, been pluralistic in its reasoning.
Collective Action and Individual Choice
The Supreme Court’s penchant for pluralism arises in two ways.
The first arises via the Court’s nature as a multimember institution
that decides cases by majority vote. The Court is a group of different
individuals appointed by different presidents and possessing of
different views about the law. It operates through collective action,
which means that for an interpretive theory to predominate, it must
consistently win the allegiance of at least five justices. And if the
theory is to have staying power, it needs continued support even as
some justices leave the bench and others join.
The need for buy-in from multiple justices decreases the likelihood
that any given constitutional theory will predominate in the Court’s
caselaw, particularly over a sustained period of time. This is not to say
it is impossible for five (or more) sitting justices to agree about the
proper theory of constitutional interpretation. Still, given the rarity
with which seats on the Court open up, it is quite the challenge to
assemble five justices who share the same philosophy. When we factor
in the divisiveness of American politics and the probability of shifts in
the balance of political power—including the power of judicial
appointment—sustained agreement over time is even more difficult to
come by.
The nature of the Supreme Court as a multimember institution is
not the only driver of interpretive pluralism. Pluralism can also
emerge as the preferred approach of an individual justice. Debates over
constitutional law include differences of opinion about whether judges
should be applying interpretive theory at all. For example, Judge J.
29
Harvie Wilkinson recently characterized leading constitutional
theories as little more than “competing schools of liberal and
conservative judicial activism.” 30 Skepticism about interpretive theory
reaches all the way to the Supreme Court. The experience of John
Roberts is illustrative. During his confirmation hearings in 2005, the
soon-to-be Chief Justice noted that rather than drawing on abstract
theory, he favors “bottom up” judging. As he explained in a response to
Senator Orrin Hatch:
If the phrase in the Constitution says two-thirds of the Senate,
everybody’s a literalist when they interpret that. Other
phrases in the Constitution are broader, [such as]
“unreasonable searches and seizures.” You can look at that
wording all day and it’s not going to give you much progress in
deciding whether a particular search is reasonable or not. You
have to begin looking at the cases and the precedents, what the
Framers had in mind when they drafted that provision.
So, yes, it does depend upon the nature of the case before you[,]
I think. 31
Chief Justice Roberts’s testimony helps to highlight why the Court’s
interpretive pluralism is not solely the product of its status as a
multimember institution. It is also the result of individual choice. Nor
is Chief Justice Roberts alone in his preference for pluralism. Five
years after his confirmation, Justice Elena Kagan offered her own
endorsement of a “case-by-case” approach to determining which
sources and arguments are relevant in the context of a particular
dispute. 32 For case-by-case justices, interpretive pluralism is an
individual phenomenon as much as an institutional one.
In defining the role of precedent, opting to decide cases in a
pluralistic, case-by-case manner is just as resonant as aligning oneself
with a particular school of interpretation. As Adrian Vermeule points
out, opting to “take each case as it comes . . . constitutes an implicit
J. HARVIE WILKINSON, III, COSMIC CONSTITUTIONAL THEORY: WHY AMERICANS ARE
LOSING THEIR INALIENABLE RIGHT TO SELF-GOVERNANCE 4 (2012). For Judge
Wilkinson, the “highest virtues of judging” lie in overcoming theory and being guided
instead by “self-denial and restraint.” Id. at 116.
31 109th Cong. 159 (2005).
32 111th Cong. 81 (2010).
30
30
choice of interpretive method and an implicit allocation of interpretive
authority. It is a choice to commit interpretation to the case-specific
discretion of the judges on the spot, as opposed to the discretion of
judges at other times and places who might formulate general
interpretive doctrine to govern the adjudicative process.” 33 Irrespective
of whether such pluralism is superior to the adoption of a discrete
interpretive theory, the two concepts play similar roles in judicial
decisionmaking.
Even so, there is a fundamental difference between interpretive
pluralism and theory-based judging when it comes to the treatment of
precedent. Pluralism denies that cases should be decided in accordance
with preexisting commitments to interpretive methodologies and
underlying normative justifications. That premise distinguishes
pluralism from other interpretive approaches. It also presents a
serious challenge when the question before the Supreme Court is
whether a flawed precedent is so problematic as to warrant overruling.
By design, judicial pluralism avoids ex ante prioritization of values and
methodologies. Yet prioritization is necessary if the application of stare
decisis depends—as I have claimed that it does—on conclusions about
the harmfulness of a precedent’s effects. There must be some
mechanism for determining which of those effects are relevant and
which are not.
The Challenges of Pluralism
In Chapter 2, I discussed several reasons for deferring to prior
decisions. Prominent among them is the utility of precedent in
enhancing the cohesiveness and impersonality of courts. At the level of
the Supreme Court, a strong system of precedent encourages the
justices to think and act like parts of an enduring institution.
The unifying effects of precedent are critical in a legal system
characterized by interpretive pluralism. Pluralism increases the
chances that substituting one justice for another will lead to the
application of an entirely different interpretive methodology,
particularly when the Court is closely divided. Deference to precedent
can smooth out the path of the law by preserving a stable core even as
justices come and go. Past decisions provide common ground and
33
ADRIAN VERMEULE, JUDGING UNDER UNCERTAINTY 157 (2006).
31
separate the inclination of the individual justice from the content of
the law. The aspiration is, to quote Amy Barrett, “a reasoned
conversation over time between justices—and others—who subscribe to
competing methodologies of constitutional interpretation.” 34
But precedent’s ability to unite the Court is threatened by the
prevalence of interpretive pluralism. I have argued that the
importance of replacing a flawed interpretation with a correct one
depends on a justice’s methodological choices and normative
commitments. I have made a similar claim about the definition of a
precedent’s scope of applicability. When the justices disagree about
baseline matters of interpretation, the Court’s decisions are likely be
inconsistent in their conceptions of precedent. Invocations of precedent
tend to restate interpretive disagreements instead of bridging them. In
the worst-case scenario, the ideal of constitutional law as stable and
enduring can give way to the notion that judicial identity and legal
meaning are one and the same.
None of this is to say interpretive pluralism is, all things
considered, a bad thing. Whether it would be better for U.S. legal
practice to move closer to a consensus theory of interpretation is an
intricate question, and one far beyond my purview here. I mean only to
suggest that pluralism, in both its institutional and individual
varieties, makes it more difficult to fashion a workable approach to
judicial precedent. A theory of precedent requires a consistent way of
defining decisions’ scope of applicability. It also requires a stable
metric for evaluating the relevant effects of a flawed decision.
Pluralism complicates both pursuits. Before there can be a wellfunctioning doctrine of precedent, there must be a plan for responding
to the challenges of pluralism.
Second-Best Stare Decisis
The Supreme Court’s doctrine of stare decisis encompasses several
discrete considerations, albeit considerations that are applied loosely
and flexibly. While the justices continue to disagree over the proper
treatment of precedent in individual cases, both the doctrinal structure
Amy Coney Barrett, Precedent and Jurisprudential Disagreement, 91 TEX. L. REV.
1711, 1737 (2013); see also Frederick Schauer, Precedent, 39 STAN. L. REV. 571, 600
(1987) (“Using a system of precedent to standardize decisions subordinates
dissimilarity among decisionmakers, both in appearance and in practice.”).
34
32
and the animating tension between legal continuity and legal
correctness are established features of the Court’s jurisprudence.
Neither reciting the governing rules nor acknowledging the choice
between continuity and correctness is sufficient to achieve consistency.
The threshold problem is that the value of getting the law right is a
controversial proposition. The difficulty is particularly acute within the
realm of constitutional law. Specifying the value of correct
interpretation carries profound implications for the treatment of
precedent, because it determines what is at stake in tolerating an
interpretive error.
One response is to view stare decisis through a particular
methodological lens by examining how a justice should treat precedent
if she is an originalist, or a common law constitutionalist, or a
pragmatist, and so on. This type of analysis is invaluable for
understanding how precedent operates within various methodological
schools. It helps judges and commentators who are considering a
particular methodology to evaluate its assumptions and implications.
It also informs arguments for why one methodology is superior to
others.
In the pages ahead, I am going to defend a different approach to
precedent, one that is designed for justices who disagree with each
other on matters of interpretive philosophy. The objective is to tailor
the doctrine of stare decisis to a world in which constitutional
interpretation is rife with methodological and theoretical
disagreements.
Those disagreements define what I call the second-best world of
constitutional law. The second-best world stands in contrast to an
idealized state of affairs in which the justices largely agree on the
appropriate ends and means of constitutional interpretation. Whether
or not interpretive pluralism is healthier than widespread interpretive
agreement, pluralism creates unique challenges for the treatment of
precedent. From the standpoint of developing a workable and coherent
doctrine of stare decisis, a world of pluralism is a second-best world.
In the economic and legal literature, the theory of the second best is
a tool for optimizing the performance of an imperfect system. Its
central insight is straightforward: If a system suffers from one flaw, it
33
might make sense to intentionally introduce another imperfection in
response to the first. Or, more formally:
Suppose that at least some of the conditions necessary to
produce a given ideal or first-best constitutional order fail
to hold. Even if it would be best to achieve full satisfaction
of all those conditions, it does not follow that it is best to
achieve as many as possible of the conditions, taken one
by one. Rather, multiple failures of the ideal can offset
one [another], producing a closer approximation to the
ideal at the level of the overall system. 35
Once we exit the ideal world, we cannot take for granted that the best
approach is to approximate the ideal as closely as possible. The
question is how best to proceed in the actual world where we find
ourselves. The answer may include steps that would, if viewed in
isolation, seem like deviations from the ideal state of affairs.
Second-best analysis offers important lessons for the doctrine of
stare decisis. I have argued that the prevalence of interpretive
pluralism creates challenges for the treatment of precedent that would
not exist under conditions of widespread agreement over constitutional
philosophy. We might imagine two possible responses. One option is to
leave the existing doctrine unchanged while hoping pluralism will not
be too much of a drag on its effectiveness. That approach is consistent
with the current status of precedent at the Supreme Court. The Court’s
discussions of stare decisis are sophisticated in many respects, but
they do not account for the ways in which interpretive philosophy
influences attitudes toward precedent.
The other option for responding to pluralism, and the one that I will
defend, is grounded in the theory of the second best. It resists the idea
that stare decisis should operate the same way in a system marked by
interpretive pluralism it would in a system of interpretive harmony.
Instead, stare decisis ought to take a different shape in response to
pluralism’s prevalence. For example, considerations that would be
relevant to the durability of precedent under ideal conditions might be
excluded from the calculus when pluralism is the order of the day. This
suggestion may seem counterintuitive, because it entails countering
35
ADRIAN VERMEULE, THE SYSTEM OF THE CONSTITUTION 10 (2011).
34
one flaw in the system by consciously introducing another. But that is
the essence of second-best thinking: When part of a system deviates
from its ideal state, sometimes the most promising path forward
entails making other adjustments instead of acting as if the initial flaw
doesn’t exist.
Second-best thinking about interpretive theory is nothing new.
Leading constitutional scholars have examined the interplay between
second-best analysis and legal interpretation, and I will draw on their
work in thinking specifically about a second-best theory of judicial
precedent. The central issue I will engage is how the Supreme Court’s
doctrine of stare decisis could be revised to operate more effectively
against a backdrop of pluralism.
The second-best theory of stare decisis uses compensating
adjustments to allow precedent to serve its purposes even in a
pluralistic environment. Without a second-best accommodation,
applications of precedent tend to reflect deeper methodological and
normative commitments. In many cases these commitments remain
submerged, but occasionally they bubble up to the surface. In 2003, the
Supreme Court concluded that retaining a precedent that failed to
protect same-sex relationships would “demean[] the lives of
homosexual persons.” 36 Years later, the Court warned that withholding
constitutional protection from corporate electioneering would validate
a “brooding governmental power” at odds with “confidence and
stability in civic discourse.” 37 The same case produced a concurrence
that chronicled various problems the Court has alleviated through its
willingness to depart from precedent. 38 Included on the list was the
Court’s most heralded reversal-of-course—its repudiation of racial
segregation in Brown v. Board of Education (1954) 39 —which is so
widely lauded precisely because the justices sought to eradicate an
insidious harm. These examples demonstrate how arguments for
overruling are grounded in conclusions about a given precedent’s
conceptual and normative flaws. Those conclusions, in turn, reflect
choices about the types of effects that are relevant to a precedent’s
retention or dismissal.
Lawrence, 539 U.S. at 575.
Citizens United, 558 U.S. at 349.
38 See id. at 377 (Roberts, C.J., concurring).
39 347 U.S. 483.
36
37
35
Under conditions of interpretive harmony, determining the
relevance of legal harms would pose little problem for the doctrine of
stare decisis. Given their agreement about interpretive theory, the
justices would possess a uniform metric for evaluating precedents’
effects. Every justice would deem precedents harmful based on some
prespecified and universal criterion, be it morality, popular sovereignty,
welfare maximization, or otherwise. The justices might not always
reach the same conclusions, but they would speak the same language.
In the real world of interpretive pluralism, the calculus changes.
The proliferation of competing methodologies makes reaching
agreement about the relevance of legal harms more difficult. The
prospect of agreement is slimmer still when the Supreme Court is
viewed as an enduring institution whose composition changes over
time. By tethering a decision’s continued vitality to the perceived
gravity of its offenses—a perception that will vary from justice to
justice—the prevailing approach to stare decisis robs precedents of
independent value beyond their attractiveness on the merits. The same
phenomena of interpretive pluralism and “reasonable disagreement” 40
that threaten the durability of particular decisions end up
destabilizing stare decisis itself.
The foregoing discussion relates to the degree of deference a
precedent commands. Problems of pluralism also affect the inquiry into
precedents’ scope of applicability. Though precedential scope is often
described in terms of binding holdings and dispensable dicta, we have
seen that the situation on the ground is more complex. Divisive
questions include whether a precedent should receive deference for its
articulated rule, what to make of judicial asides, and how future
justices should treat an opinion’s expression of its rationale. The
answers depend on underlying assumptions about the competence,
institutional role, and constitutional authority of the courts.
Establishing a consistent account of precedential scope requires
doctrinal accommodations that can overcome interpretive disputes.
Before we can pursue a doctrine of precedent that makes good on its
promise of uniting judges across time and enhancing the impersonality
of constitutional law, there must be a reconceptualization to account
See RICHARD H. FALLON, JR., IMPLEMENTING THE CONSTITUTION 103 (2001) (“Stare
decisis . . . furnishes a functionally crucial response to the phenomenon of reasonable
disagreement.”).
40
36
for the pervasiveness of interpretive disputes. Our second-best world of
interpretive disagreement requires a second-best doctrine of stare
decisis.
Under the second-best approach, precedent becomes a tool of
unification rather than division. It represents a shared commitment to
respecting the Court’s institutional past. This commitment sometimes
calls upon the Court to uphold a precedent notwithstanding its
deleterious (in the view of today’s justices) effect. Such deference
carries a cost; the conscious perpetuation of error should not occur
lightly. But the question is not whether stare decisis has a price. It is
whether the price is worth paying. I will contend that the virtues of
deference often justify the toleration of error, so long as stare decisis is
applied in a manner that is consistent, impersonal, and—to the
greatest extent possible—independent of disputes over interpretive
philosophy. Indeed, quotidian disputes over constitutional methodology
are a leading reason why stare decisis is so important. Without
deference, a change in the composition of the Court is a change in the
fabric of the law. The driving objective of second-best stare decisis is
the separation of precedent from interpretive theory in a way that
transcends the identities and proclivities of individual judges.
Toward a Second-Best Approach
The ensuing Chapters develop the theory of second-best stare
decisis. Consistent with my emphasis on the value of continuity, I seek
to preserve the existing doctrine of stare decisis to the greatest extent
possible. At the same time, the second-best approach identifies
revisions and accommodations designed to insulate stare decisis from
disputes over interpretive philosophy.
On the issue of precedential strength, second-best analysis yields
two potential strategies for addressing the challenges of pluralism. The
first solution, which I explain in Chapter 6, focuses on reconstructing
the doctrine of stare decisis around considerations that can operate
independently of interpretive philosophy. Familiar considerations such
as procedural workability, factual accuracy, and reliance expectations
can be redefined to reduce their dependence on constitutional theory,
making them suitable for application by judges across the
methodological spectrum. At the same time, some factors that might be
relevant under conditions of interpretive harmony—including
37
jurisprudential coherence, flagrancy of error, and a precedent’s
perceived harmfulness—cannot function against the backdrop of
pluralism. Second-best stare decisis generally excludes those factors
from the examination of a precedent’s fitness for retention.
In developing my account of second-best stare decisis, I begin by
taking for granted that a majority (at least) of Supreme Court justices
continues to support stare decisis as a general matter. This
assumption is based on the justices’ statements in case after case about
the centrality of precedent and the virtues of deference. The justices
commonly describe stare decisis as a mechanism for “maintaining
public faith in the judiciary as a source of impersonal and reasoned
judgments.” 41 They depict the doctrine as integral to “the very concept
of the rule of law underlying our own Constitution,” which “requires
such continuity over time that a respect for precedent is, by definition,
indispensable.” 42 They recognize precedent as promoting “the
evenhanded, predictable, and consistent development of legal
principles,” while at the same time “foster[ing] reliance on judicial
decisions” and “contribut[ing] to the actual and perceived integrity of
the judicial process.” 43 Deference to precedent “permits society to
presume that bedrock principles are founded in the law rather in the
proclivities of individuals,” an effect that “contributes to the integrity
of our constitutional system of government.” 44 Indeed, “[f]idelity to
precedent” is “vital to the proper exercise of the judicial function.” 45 It
is nothing less than a “foundation stone of the rule of law.” 46 As these
statements illustrate, the doctrine of stare decisis enjoys strong
support at the Supreme Court. The controversy surrounds its
application to particular cases.
Nevertheless, I briefly relax the assumption of a Court that
remains committed to stare decisis. My inquiry shifts to why an
individual justice should accept second-best stare decisis without any
guarantee that her present and future judicial colleagues will follow
suit. I defend the second-best theory as a valuable tool for promoting
continuity, impersonality, and coordinated action, and I explain why
Moragne v. States Marine Lines, Inc., 398 U.S. 375, 403 (1970).
Casey, 505 U.S. at 854.
43 Payne v. Tennessee, 501 U.S. 808, 827 (1991).
44 Vasquez v. Hillery, 474 U.S. 254, 265 (1986).
45 Citizens United, 558 U.S. at 377 (2010) (Roberts, C.J., concurring).
46 Michigan v. Bay Mills Indian Community, 134 S. Ct. 2024, 2036 (2014).
41
42
38
an individual justice’s support for the theory should not depend on the
behavior of her peers or successors on the Court.
As an alternative to the doctrinal approach to precedential strength,
Chapter 7 describes a structural response to interpretive pluralism
that requires overrulings to receive support from a supermajority of
justices. The structural response increases the likelihood that
overrulings will rest on factors that bridge methodological divides; the
greater the number of votes required to achieve a particular result, the
better the chances that some votes must come from justices who
disagree with one another on matters of interpretive theory. Unlike the
doctrinal version of second-best stare decisis, the supermajority
requirement does not promote judicial impersonality on the individual
level by encouraging a justice to subordinate her views to those of her
predecessors. I will emphasize this distinction in depicting the
doctrinal approach to precedential strength as superior to its
structural cousin. Nevertheless, the simplicity of the supermajority
requirement makes it worthy of consideration by those who see value
in legal continuity but who have reservations about the efficacy or
wisdom of altering the content of stare decisis doctrine.
Second-best stare decisis also suggests a revised approach to
defining a precedent’s scope of applicability, as I explain in Chapter 8.
Existing practice supports the principle that rules announced by the
Supreme Court warrant deference in future cases. Much the same is
true of the principle that judicial asides and hypotheticals do not
warrant deference. Given their widespread acceptance, these norms
can guide determinations of precedential scope in the second-best
world. As for statements of rationale contained in Supreme Court
decisions, second-best stare decisis counsels an intermediate position
that treats such statements as worthy of deference, but only to the
extent they illuminate a precedent’s rule of decision.
Stepping back, my basic claim is that for stare decisis to do what
the justices have said it should do, the existing doctrine must be
reconsidered. Most importantly, analysis of the strength and scope of
precedent must be separated from debates about interpretive
philosophy, even if those debates would be relevant to the treatment of
precedent under ideal conditions. Instead of linking precedent to
competing interpretive theories, the second-best approach seeks to
39
infuse stare decisis with independent content that holds steady
regardless of a justice’s methodological or normative inclinations.
When a justice defers to her predecessors despite reservations
about the merits of their decisions, she reinforces the generality of
constitutional law. Fidelity to precedent can “transform[] the Court
from an ever-changing collection of individual judges to an institution
capable of building a continuing body of law.” 47 It also underscores the
distinction between the judiciary and the political branches, promoting
a vision in which changing the law and changing the judge are very
different things.
Disagreements over interpretive theory are here to stay. Outright
disavowals of theory also appear to be sturdy components of the
constitutional landscape. But stare decisis need not be undermined by
such phenomena. Untangled from debates over interpretive theory,
second-best stare decisis embodies a fundamental commitment to
continuity and impersonality—a commitment to the rule of law rather
than the rule of men and women. It contributes to the perception and
reality of a Supreme Court that speaks “for the Constitution itself
rather than simply for five or more lawyers in black robes.” 48
Daniel A. Farber, The Rule of Law and the Law of Precedents, 90 MINN. L. REV.
1173, 1183 (2006).
48 Earl M. Maltz, Some Thoughts on the Death of Stare Decisis in Constitutional Law,
1980 WIS. L. REV. 467, 484.
47
40
CHAPTER 6: PRECEDENTIAL STRENGTH IN DOCTRINAL PERSPECTIVE
This Chapter considers the implications of second-best stare decisis
for determinations of precedential strength. The first step is examining
how the Supreme Court’s existing doctrine of stare decisis operates in
a world of interpretive pluralism. From there, I suggest a series of
doctrinal revisions to overcome the challenges of pluralism. I seek to
preserve and refine components of the existing doctrine that provide
meaningful guidance in a second-best world of pluralism, while
limiting or excluding components whose content depends on
methodological and normative commitments that vary from judge to
judge. Notwithstanding other benefits that can arise from a system of
precedent-based judging—benefits such as predictability, efficiency,
and the protection of expectations—the core of second-best stare decisis
is impersonality and continuity. Precedent allows the law to transcend
the moment.
Second-best stare decisis leaves individual justices at liberty to
apply their preferred methods of interpretation in cases of first
impression and in diagnosing whether a precedent was erroneous. But
it revises the Court’s rules of precedent to pursue a doctrine that works
across methodological lines. In doing so, it facilitates coordinated
action among justices who are inclined to view the world differently.
This latter feature most clearly separates a commitment to secondbest stare decisis from a commitment to interpretive methodologies
such as originalism or living constitutionalism. Fostering agreement
around a particular interpretive methodology is a tall order in our
pluralistic legal culture. It requires some justices to take the dramatic
step of disavowing their constitutional philosophies. By contrast,
asking an originalist or living constitutionalist to recognize a
meaningful role for precedent does not require her to abandon her
interpretive philosophy wholesale. Second-best stare decisis reflects a
shared sacrifice, because it calls upon every justice to give presumptive
respect to some decisions—and, by implication, to the methodologies
and values that yielded them—that she views as flawed. The
endorsement of second-best stare decisis is more egalitarian, and less
jarring, than the displacement of one interpretive methodology by
another.
The Doctrine of Stare Decisis
41
Stare decisis is founded on the principle that today’s judges do not
write on a clean slate. That same principle applies to the doctrine of
stare decisis itself. Precedent is an integral component of American
legal practice. Its importance reaches back to the founding, and beyond.
The resonance of stare decisis is apparent in Alexander Hamilton’s
characterization of precedent as a safeguard against the exercise of
“arbitrary discretion in the courts.” 49 It is equally evident in James
Madison’s view that the Constitution’s ambiguities would be
“liquidated” over time through judicial and political decisions. 50
Within the Supreme Court’s caselaw, stare decisis has gone beyond
abstract principles to yield a discrete set of relevant considerations.
When a new justice takes her place on the bench, she is greeted by a
preexisting doctrine of stare decisis, just as she encounters preexisting
doctrines on scores of other topics. The Court has described several
factors as relevant to the analysis. The most prominent account comes
from Planned Parenthood of Southeastern Pennsylvania v. Casey
(1992), where the Court discussed its decision in Roe v. Wade (1973)
regarding the constitutional status of abortion rights. In surveying the
components of stare decisis doctrine, the Casey majority identified the
following factors:
So in this case we may enquire whether Roe’s central rule
has been found unworkable; whether the rule’s limitation on
state power could be removed without serious inequity to
those who have relied upon it or significant damage to the
stability of the society governed by it; whether the law’s
growth in the intervening years has left Roe’s central rule a
doctrinal anachronism discounted by society; and whether
Roe’s premises of fact have so far changed in the ensuing two
THE FEDERALIST NO. 78 (Alexander Hamilton).
See, e.g., THE FEDERALIST NO. 37 (James Madison) (“All new laws, though penned
with the greatest technical skill, and passed on the fullest and most mature
deliberation, are considered as more or less obscure and equivocal, until their
meaning be liquidated and ascertained by a series of particular discussions and
adjudications.”); Caleb Nelson, Stare Decisis and Demonstrably Erroneous Precedents,
87 VA. L. REV. 1, 10–14 (2001) (explaining and contextualizing Madison’s position); cf.
THE FEDERALIST NO. 78 (A. Hamilton) (noting that in the face of inconsistent laws, “it
is the province of the courts to liquidate and fix their meaning and operation”).
49
50
42
decades as to render its central holding somehow irrelevant
or unjustifiable in dealing with the issue it addressed. 51
In developing a second-best approach to precedent, it will be useful
to begin with the factors included in Casey’s influential account. For
ease of exposition, I have reordered them and described them as
procedural workability, factual accuracy, jurisprudential coherence,
and reliance and disruption. After discussing the suitability of each
factor in our second-best world, I extend the analysis to additional
considerations that often play a role in the Supreme Court’s
applications of stare decisis. Among the most important of those
considerations are whether the perceived severity of a precedent’s
error should affect its retention, and how to account for a precedent’s
negative consequences.
Procedural Workability
The inquiry into workability responds to the “mischievous
consequences to litigants and courts” that can result from a vague or
byzantine rule of decision. 52 The rationale for paying attention to
workability is intuitive. Precedents that have created “uncertainty and
arbitrariness of adjudication” warrant reconsideration. 53 Those
precedents impose costs that all judges can recognize as undesirable.
The characterization of a decision as workable or unworkable
sometimes seems to track views about the decision’s soundness on the
merits. 54 This correlation might suggest the diagnosis of workability
cannot be separated from a justice’s interpretive and normative
preferences. And, in fact, there is little doubt that when workability is
51 Casey, 505 U.S. at 855; see also Montejo v. Louisiana, 556 U.S. 778, 792–93 (2009);
Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877, 923–29 (2007)
(Breyer, J., dissenting); Barry Friedman, The Wages of Stealth Overruling (with
Particular Attention to Miranda v. Arizona), 99 GEO. L.J. 1, 26 (2010); Michael Stokes
Paulsen, Does the Supreme Court’s Current Doctrine of Stare Decisis Require
Adherence to the Supreme Court’s Current Doctrine of Stare Decisis?, 86 N.C. L. REV.
1165, 1173–98 (2008).
52 Swift & Co. v. Wickham, 382 U.S. 111, 116 (1965).
53 Johnson v. United States, 135 S. Ct. 2551, 2562 (2015); see also Thomas R. Lee,
Stare Decisis in Economic Perspective: An Economic Analysis of the Supreme Court’s
Doctrine of Precedent, 78 N.C. L. REV. 643, 670 (2000).
54 Compare, for example, Dickerson v. United States, 530 U.S. 428, 444 (2000)
(defending the relative workability of the Miranda rule), with id. at 463 (Scalia, J.,
dissenting) (disputing Miranda’s “supposed workability”).
43
defined too broadly, methodological and normative commitments can
creep into the analysis. At base, however, a decision’s workability is
independent of disputes over interpretive philosophy. The workability
analysis is amenable to principled application regardless of a
particular justice’s interpretive predilections. It simply requires a
reconceptualization.
The critical step is rejecting the premise that a precedent becomes
unworkable because a justice disagrees with its rationale or is troubled
by its results. Saying a precedent is poorly reasoned or has wrought
moral or practical harms is not an argument from workability. It is an
argument about the precedent’s interpretive approach and substantive
effects.
The proper reasons for paying attention to a decision’s workability
are procedural in nature. They deal with whether courts, litigants, and
other stakeholders have been able to understand and apply a rule
without undue difficulty. A rule of decision that is hopelessly
convoluted or exceedingly vague renders a precedent unworkable
regardless of its rationale and substantive effects. Likewise, a rule of
decision that is unmistakably clear must be acknowledged as
procedurally workable even if its substantive effects have been
disastrous.
I will discuss the treatment of substantive effects in greater depth
below. For now, the key is distinguishing substantive effects from
considerations of procedural workability, the latter of which possess
independent force regardless of a justice’s interpretive methodology.
Whether a precedent has been clear enough for courts to understand
and apply does not depend on whether a particular justice is an
originalist, a pragmatist, or a common law constitutionalist. That
makes procedural workability an appropriate consideration for the
doctrine of stare decisis in our second-best world of interpretive
pluralism.
This is not to say the justices will always agree. There are different
degrees of workability, and disputes will surely arise about whether a
particular precedent is—notwithstanding some procedural flaws—
capable of furnishing adequate guidance to the bench and bar. Secondbest stare decisis is compatible with that reality. The second-best
approach is not quixotic. It does not seek to eliminate disagreements in
44
judgment. The goal is to facilitate reasoned deliberation in a common
grammar that transcends interpretive disputes. Second-best stare
decisis can tolerate competing conclusions in the context of an
individual case. What it rejects are criteria that have no independent
content unless they are situated within a particular methodological
framework. Because the creation of workable decisions—and the
revision or eradication of unworkable ones—has value across a range
of methodological perspectives, procedural workability stands as a
legitimate component of second-best stare decisis.
Factual Accuracy
Judicial decisions contain factual premises, and those premises can
be wrong. The error may have existed from the beginning, as when the
Supreme Court repeats a mistaken statement by a litigant regarding
federal immigration policy. 55 Or a premise may have disintegrated
over time, as when technological developments undermine the Court’s
prior characterization of certain forms of media. 56 In either situation,
factual accuracy is compromised.
As with their treatment of workability, courts occasionally conflate
diagnoses of factual error with assessments of a precedent’s legal
reasoning. Consider the Supreme Court’s discussion of factual
mistakes in Casey. The Court concluded there had been no factual
developments that undermined the “central holding” of Roe regarding
viability as the critical point for determining the government’s power
to prohibit nontherapeutic abortions. 57 In explaining its conclusion, the
Court invoked two precedents that it characterized as resting on
factual mistakes. The first case was Lochner v. New York (1905), which
Casey described as reflecting inaccurate assumptions about “the
capacity of a relatively unregulated market to satisfy minimal levels of
human welfare.” 58 The second case was Plessy v. Ferguson (1896),
which Casey described as overlooking the pernicious stigmatization of
racial segregation. 59
The example is discussed in Allison Orr Larsen, Factual Precedents, 162 U. PA. L.
REV. 59, 63–64 (2013).
56 FCC v. Fox Television Stations, Inc., 556 U.S. 502, 533 (2009) (Thomas, J.,
concurring).
57 Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 860
(1992).
58 Id. at 861–62.
59 See id. at 863.
55
45
The Casey Court’s understanding of what constitutes a factual
premise is too capacious to be useful. One can accept that both Lochner
and Plessy deserved to be overruled while still recognizing that their
respective errors ranged beyond their factual premises. The dispositive
change from Plessy to Brown v. Board of Education (1954), 60 and from
Lochner to West Coast Hotel Co. v. Parrish (1937), 61 was not empirical
reality. It was the opinions and values through which reality is
perceived and understood. Opinions and values are vital, but they do
not possess the objectivity of facts.
The Casey approach is not the only way to identify which of a
judicial decision’s components are factual. Factual content can be
understood more narrowly as driven by empirical observations that do
not depend on methodological or normative commitments. A useful
example comes from the field of broadcast regulation. The Supreme
Court ruled in FCC v. Pacifica Foundation (1978) that broadcasters
are subject to punishment for disseminating indecent speech. Among
the reasons for this conclusion were broadcast media’s pervasiveness
and accessibility to children. 62
Pacifica was controversial from the outset, and it remains so today.
Challenges have made their way to the Supreme Court twice in recent
years. Both times, the Court disposed of the cases without overruling
or reaffirming Pacifica. But Justice Thomas and Justice Ginsburg
penned separate opinions renouncing the Pacifica approach. Justice
Ginsburg was brief in her remarks, arguing that Pacifica was
mistaken from the beginning and had become worse in light of
changing technologies. 63 Justice Thomas offered more elaboration
along similar lines. He criticized Pacifica and a predecessor case, Red
Lion Broadcasting Co. v. FCC (1969), as a “deep intrusion into the
First Amendment rights of broadcasters.” He also described
technological advances as undermining the idea that broadcasting is
uniquely pervasive and therefore more susceptible to regulation. 64
347 U.S. 483.
300 U.S. 379.
62 See 438 U.S. 726, 748–51.
63 FCC v. Fox Television Stations, Inc., 132 S. Ct. 2307, 2321 (2012) (Ginsburg, J.,
concurring in the judgment).
64 Fox Television Stations, 556 U.S. at 531–34 (Thomas, J., concurring).
60
61
46
Whatever the merits of these criticisms, the question for purposes
of stare decisis is what happens if a majority of the Court follows the
lead of Justices Thomas and Ginsburg and concludes that Pacifica is
wrong. Pacifica exemplifies a case whose factual premises have eroded.
It is no longer true to say broadcasters are unique in their
pervasiveness or accessibility to children. Because those characteristics
were important components of Pacifica’s rationale, the decision is
properly viewed as subject to reconsideration. The presumptive
deference owed to Pacifica has been rebutted by the disintegration of
its factual predicates. That does not necessarily mean Pacifica should
be overruled. There may be other reasons for preserving it (for instance,
based on the reliance it has engendered) or possibilities for
reconceptualizing it (for instance, by focusing on the government’s
authority to license broadcasters). Even so, flawed factual premises
warrant a fresh look at the decision.
Another example comes from the field of taxation. The Supreme
Court addressed the tax obligations of mail-order sellers in Quill Corp.
v. North Dakota (1992). 65 The case dealt with a state’s authority to
impose tax-collection obligations on sellers who maintained no physical
presence within the state. Viewed through the lens of precedent, the
proper outcome seemed clear enough. The Court’s prior opinion in
National Bellas Hess, Inc. v. Department of Revenue (1967) had sided
with sellers in comparable circumstances. 66 But in the years since
Bellas Hess was decided, developments in related areas of
constitutional law had arguably rendered it a doctrinal outlier. The
question in Quill was whether the Court should reaffirm Bellas Hess
despite doubts about its soundness by contemporary standards or
overrule the case to enhance the internal consistency of constitutional
law.
The Court chose the former course. It reaffirmed that the
Commerce Clause forbids a state from imposing tax-collection
obligations on mail-order sellers who lack a physical presence in the
state. Along with defending Bellas Hess’s application of the Commerce
Clause as suited to its particular context, the Court noted the
importance of reliance expectations. It explained that mail-order
sellers had expended significant resources based on the Court’s
65
66
504 U.S. 298.
386 U.S. 753.
47
previous decision and that altering the rules could threaten “the basic
framework of a sizable industry.” 67
Today’s retail world is different from the one the Court encountered
in Quill. As Justice Kennedy recently noted, the years since Quill’s
issuance have witnessed “dramatic technological and social changes”
that arguably give many out-of-state sellers “a sufficiently ‘substantial
nexus’” to justify imposing tax-collection obligations upon them. 68
According to Justice Kennedy, developments in e-commerce have
changed what it means for a business to be “present” in a state.
Like the innovations in the media since Pacifica, the technological
developments in online retailing since Quill are the types of bona fide
factual changes that rebut the presumption of deference to precedent
even in a second-best world of interpretive pluralism. The Court’s prior
opinions considered the degree to which sellers can connect with a
given state notwithstanding the lack of any bricks-and-mortar
presence. The proliferation of Internet retail has changed that calculus
considerably. Some justices might nevertheless conclude out-of-state
sellers should remain exempt from tax- obligations. But for justices
who reject that conclusion, changes to the retail industry brought
about by technological advances are sufficient to justify the
reconsideration of precedent. Like the broadcast media, interstate
retailers operate in an environment that has undergone significant
changes. The evolution of the retail environment is a sensible reason
for reconsidering precedent, and one that is not bound up with any
particular methodology of interpretation. That makes it an appropriate
consideration within a system of second-best stare decisis.
Again, it does not necessarily follow that Pacifica and Quill should
be overruled, or that Lochner or Plessy shouldn’t have been. Whether
an overruling is warranted depends on other factors as well. We will
reach these additional factors in due course. My present aim is to
illustrate how the concept of factual inaccuracy can be pared down to
its objective core. A streamlined conception of what constitutes a
factual premise facilitates deliberation across methodological lines. An
incorrect statement of fact is an incorrect statement of fact, regardless
of a justice’s interpretive philosophy.
Quill, 504 U.S. at 317.
Direct Marketing Association v. Brohl, 135 S. Ct. 1124, 1135 (2015) (Kennedy, J.,
concurring).
67
68
48
Jurisprudential Coherence
Under existing law, precedents lose their force if they become “mere
survivor[s] of obsolete constitutional thinking.” 69 The operative
question, to recall the Supreme Court’s language in Casey, is “whether
the law’s growth in the intervening years” has left a precedent as “a
doctrinal anachronism discounted by society.” 70
In theory, jurisprudential coherence is another consideration that
can be separated from methodological disagreements. The consistency
of one precedent with another does not depend on any particular
interpretive philosophy. It reflects the belief that sound judging aspires
to consistent decisions—a commitment shared by justices across the
methodological spectrum. But while decisions will sometimes be
glaringly incompatible, in other cases the inquiry into jurisprudential
coherence runs parallel with debates over a precedent’s soundness.
Was the Supreme Court’s pre-Citizens United caselaw inconsistent in
failing to protect corporate electioneering notwithstanding the strong
protection given to political speech more generally, or did the caselaw
properly recognize the unique dynamics of corporate advocacy in
elections? Is the Court’s brightline rule regarding the tax liability of
out-of-state retailers an anachronistic holdover from an absolutist era,
or a specialized application that makes sense within the broader
doctrinal scheme? How did the Court’s cases involving abortion rights
and equal protection affect the durability of its earlier decision to
uphold the constitutionality of a criminal prohibition against same-sex
relationships? Without an objective baseline, there is too great a risk of
these questions being answered by reference to interpretive
commitments that vary from justice to justice, robbing stare decisis of
its ability to serve as a bridge between methodologies.
The problem is exacerbated by inconsistency in defining precedents’
scope of applicability. If we lack a clear sense of what a precedent
stands for, we cannot determine its compatibility with other cases. But
even a uniform definition of precedential scope would leave a great
deal of ambiguity in the assessment of jurisprudential coherence.
When the Court was deliberating about Citizens United, there was no
Casey, 505 U.S. at 857; see also MICHAEL J. GERHARDT, THE POWER OF PRECEDENT
31 (2008).
70 Casey, 505 U.S. at 855.
69
49
doubt that its precedents permitted restrictions on corporate
electioneering. The question was whether other cases, while not
directly applicable, pointed in a different direction. Likewise, when the
Court reconsidered the constitutionality of prohibitions against samesex relationships in Lawrence v. Texas (2003), Bowers v. Hardwick
(1986) was understood to be the governing law. That did not resolve
the issue of jurisprudential coherence; the Court also needed to
evaluate the impact of other decisions that illuminated broader
principles in its caselaw despite not being squarely on point.
If definitions of precedential scope do not settle debates about
jurisprudential coherence, other considerations must be doing the work.
To be sure, a justice can consider a series of cases, extract their general
themes, and ask whether a given precedent deviates from them. The
issue will always be whether outliers are justified on their own terms.
There might be good reasons for treating corporate electioneering
differently from electioneering by individuals, or for recognizing special
rules for the tax treatment of out-of-state retailers. Or there might not.
These determinations are complicated and contestable.
The best explanation for why Citizens United treated the Court’s
precedents on corporate electioneering as anomalous (and wrong)
rather than context-sensitive (and right) is the majority’s conclusion
that the prior decisions were antithetical to core First Amendment
values. The same is true of Lawrence. The majority’s characterization
of Bowers as a doctrinal anachronism was connected to the conclusion
that Bowers permitted states to “demean” the “existence” of consenting
adults and to “control their destiny by making their private sexual
conduct a crime”—a point the Lawrence majority underscored by
stating Bowers “was not correct when it was decided.” 71 The foregoing
depictions are emblematic of a more general phenomenon in the
Court’s caselaw. They are steeped in methodological and normative
commitments—the types of commitments second-best stare decisis
seeks to avoid.
As I have noted, factors such as procedural workability and factual
accuracy can also intermingle with interpretive philosophy. But those
factors are more susceptible to a paring down that separates them
from methodological and normative commitments. The inquiry into
71
539 U.S. 558, 578.
50
jurisprudential coherence admits of no such narrowing, at least in
practical terms. Its wide-ranging nature impedes its decoupling from
interpretive and normative debates. Notwithstanding its potential
relevance in a first-best world of interpretive agreement, the inquiry
into jurisprudential coherence must be excluded from second-best stare
decisis.
Reliance and Disruption
Procedural unworkability, factual inaccuracy, and jurisprudential
incoherence are potential reasons for rejecting a flawed precedent. On
the other side of the scale is the impact of an overruling on reliance
expectations. The Supreme Court has expressed reluctance about
overruling precedents that command significant reliance. 72
While disputes over the extent of reliance are significant and
unavoidable, they need not collapse into debates about interpretive
philosophy. The assessment of reliance is fundamentally an empirical
undertaking, albeit one in which the data will never be complete.
There is no reason why a justice’s assessment of the reliance a
precedent has engendered should vary depending on whether she is an
originalist, a living constitutionalist, or otherwise. Despite its
complexity, the reliance analysis is an objective inquiry in which a
justice’s interpretive predilections do not dictate her conclusions about
the degree of disruption an overruling will cause. Reliance accordingly
is well-suited to second-best stare decisis.
The difficulty is determining which types of reliance are relevant.
The Supreme Court’s caselaw suggests disruption is most problematic
in situations involving investment-backed expectations and rights in
property and contract. 73 Nevertheless, the rationale for protecting
reliance can apply to noneconomic liberties as well. Whatever the
nature of the underlying right, when stakeholders have taken tangible
For recent expressions of this point, see Michigan v. Bay Mills Indian Community,
134 S. Ct. 2024, 2036 (2014); Harris v. Quinn, 134 S. Ct. 2618, 2652 (2014) (Kagan, J.,
dissenting). Cf. Walton v. Arizona, 497 U.S. 639, 673 (1990) (Scalia, J., concurring in
part and concurring in the judgment). See also Henry P. Monaghan, Taking Supreme
Court Opinions Seriously, 39 MD. L. REV. 1, 7 (1979) (““History has its claims, at
least where settled expectations of the body politic have clustered around
constitutional doctrine.”).
73 Casey, 505 U.S. at 855–56; Quill, 504 U.S. at 317.
72
51
steps or made concrete plans in reliance on Supreme Court
pronouncements, the disappointment of their expectations is pertinent.
Reliance by governmental officials such as legislators presents a
more complicated issue. Some Supreme Court decisions treat
governmental reliance as a legitimate part of the stare decisis
calculus. 74 A potential counterpoint is Citizens United v. FEC (2010),
which overruled a precedent that had commanded considerable
reliance from Congress and state legislatures. But the Citizens United
majority did not declare legislative reliance to be inapposite; rather, it
described that factor as “not . . . compelling.” 75 In light of the Court’s
other opinions giving regard to legislative reliance, it would be curious
if the Citizens United majority meant to sweep away its past practice
without acknowledging what it was doing. The “not . . . compelling”
language is better understood to mean that within the context of
corporate electioneering, legislative reliance was not sufficiently
weighty to save a flawed precedent.
By treating government reliance as meaningful, courts express
respect for good-faith efforts by coordinate branches to comply with
judicial edicts. Protecting legislative reliance also acknowledges the
reality that it is private citizens who bear the costs of sending
politicians back to the drawing board when judges change the rules of
the game. These considerations suggest governmental reliance is
relevant to the stare decisis calculus even in the second-best world of
interpretive pluralism.
The same cannot be said for reliance by society at large. Broad
notions of societal reliance on precedent have played a role in major
constitutional cases. In Casey, the Court recognized the interests of
“people who have ordered their thinking and living around” the
continued vitality of Roe. 76 Similarly, when it reaffirmed Miranda v.
Arizona (1966), the Court cited the status of the Miranda warnings as
“part of our national culture.” 77 Roe and Miranda are exceptional in
their salience and profile, creating questions about whether the Court’s
Randall v. Sorrell, 548 U.S. 230, 244 (2006) (Breyer, J.); Harris, 536 U.S. at 567–68
(plurality op.); Hilton v. South Carolina Public Railways Commission, 502 U.S. 197,
202 (1991).
75 Citizens United, 558 U.S. at 365.
76 Casey, 505 U.S. at 856.
77 Dickerson, 530 U.S. at 428.
74
52
invocations of societal reliance should extend to other areas of
constitutional law. Further, unlike assertions of private and legislative
reliance, appeals to societal reliance do not depend on the concrete
expectations of stakeholders whom an overruling will affect most
directly. The form of analysis is necessarily more abstract.
In addition, the objectives served by protecting societal reliance are
promoted to a considerable extent by the very existence of a
meaningful doctrine of stare decisis—including one that does not
expressly consider society’s reliance on a given precedent. Societal
reliance is distinct from private and legislative reliance because it
captures the myriad effects that overruling a well-known precedent
can have on various communities, such as people who have not taken
tangible action based on the precedent. To illustrate, return to the
example of Miranda. The overruling of Miranda—and the
corresponding reinvention of the rules of engagement for criminal
suspects—could matter immensely to perceptions about the law’s
continuity and impersonality, even for those who will never hear their
Miranda rights read to them. This does not mean Miranda is
impervious to reconsideration. But it does mean an overruling would
affect more than the people most directly affected by Miranda’s rule.
The best response is to demand a special justification for overruling a
precedent—whether Miranda or any other decision—that goes beyond
disagreement with the precedent’s reasoning. A general practice of
deference to precedent is a way of respecting societal reliance on the
universe of Supreme Court pronouncements. Such a practice reduces
the need for undertaking difficult and controversial inquiries into
societal reliance within the context of specific disputes.
Societal reliance might well be a sensible consideration for a
doctrine of stare decisis fashioned under first-best conditions of abiding
interpretive agreement. In our second-best world, a doctrine of stare
decisis must make compromises in order to appeal to a wide crosssection of judges. Societal reliance does not enjoy the same grounding
in existing caselaw as do more direct forms of reliance by private
citizens and government officials. It rests on an entirely different set of
conceptual underpinnings. What is more, societal reliance can receive
indirect protection from a doctrine of stare decisis that demands a
special justification for overruling precedent. On balance, these
arguments lead me to conclude that societal reliance must be excluded
from second-best stare decisis. Though private reliance and
53
governmental reliance raise their fair share of challenges in
application, they are better established and more amenable to
consistent analysis than reliance by society at large. Expectations still
matter in the second-best world, but societal reliance must find its
protection in the presumptive respect all precedents receive.
Flawed Reasoning and Flagrancy of Error
Beyond the Casey factors, another consideration that frequently
finds its way into discussions of stare decisis is the reasoning of the
precedent under review. The presence of error is not a sufficient
ground for overruling a precedent. As the Supreme Court wrote in
Casey, “a decision to overrule should rest on some special reason over
and above the belief that a prior case was wrongly
decided.” 78Nevertheless, the justices sometimes include the soundness
of a precedent’s reasoning among the relevant considerations in
deciding whether to preserve it. 79
A way to resolve this apparent tension is by drawing distinctions
among precedents based on the flagrancy of their errors. On this
account, the fact that a precedent is wrong—or “badly reasoned,” if you
prefer—is not a reason to overrule it. By comparison, when a precedent
is not simply mistaken but clearly or manifestly so, the extent of its
error becomes a reason for doing away with it. Caleb Nelson has
provided a powerful defense of this position on both theoretical and
historical grounds. He contends that withholding deference from
manifestly erroneous precedents may be a desirable alternative to
embracing a “general presumption against overruling past decisions.” 80
Within the contours of a single interpretive school, classifying
dubious precedents based on the flagrancy of their error is sensible. If,
for example, a justice thinks constitutional rules should be assessed
pragmatically based on their costs and benefits, it is natural that the
justice will distinguish between precedents whose costs slightly
Casey, 505 U.S. at 864; see also Harris, 134 S. Ct. at 2652 (Kagan, J., dissenting)
(“The special justifications needed to reverse an opinion must go beyond
demonstrations (much less assertions) that it was wrong; that is the very point of
stare decisis.”); Caleb Nelson, Stare Decisis and Demonstrably Erroneous Precedents,
87 VA. L. REV. 1, 8 (2001).
79 See, e.g., Citizens United, 558 U.S. at 362–63; Montejo, 556 U.S. at 793; Payne, 501
U.S. at 827.
80 Nelson, Stare Decisis and Demonstrably Erroneous Precedents, supra note __, at 4.
78
54
outweigh their benefits and precedents whose costs greatly outweigh
their benefits. In this example, the latter category includes cases of
manifest error—sufficient to jeopardize a precedent’s continued
vitality—whereas the former category includes cases of ordinary error
that do not themselves furnish sufficient justification for overruling. A
similar analysis applies to other interpretive methodologies. If one
accepts the validity of originalism, it is prudent to distinguish between
precedents that reflect probable misreadings of the constitutional text
and those that reflect blatant misreadings. Again, such a distinction
facilitates the sorting of precedents into the categories of manifest
error and ordinary error. That categorization informs the
determination of which mistakes the Court should live with and which
it should take measures to rectify.
In our pluralistic world, the prospect of distinguishing precedents
based on their degree of error becomes more complicated. 81 To
illustrate, simply combine the examples discussed in the previous
paragraph. If Justice A believes that pragmatism is the proper method
of interpreting the Constitution, how is she to determine whether a
precedent that is reasoned on originalist grounds reflects an ordinary
error or a manifest error? And if Justice B adheres to originalism, how
is she to distinguish between ordinary error and manifest error in
precedents that are couched in pragmatic terms? 82
A potential solution, which is based on interpretive fidelity to one’s
preferred philosophy, is to treat as manifestly erroneous all precedents
that reflect a methodology different from one’s own. An originalist
justice would characterize all nonoriginalist interpretations as
manifestly erroneous and, thus, subject to overruling. A nonoriginalist
justice would take the same view of precedents decided on originalist
grounds. And so the divide between ordinary error and manifest error
would collapse into disagreements over interpretive philosophy. The
problem with this approach is that it sacrifices the ability of stare
decisis to unite justices who adhere to divergent theories of
Professor Nelson notes this issue, recognizing the risk that “current judges may be
committed to an entirely different interpretive method than their predecessors, and
they may be too quick to decide that their predecessors' method was illegitimate.” Id.
at 67.
82 See Jill E. Fisch, The Implications of Transition Theory for Stare Decisis, 13 J.
CONTEMP. LEGAL ISSUES 93, 101 (2003) (arguing that the outcome of a Supreme
Court case “may reflect a variety of policy, methodological and political choices, but is
unlikely to demonstrate that the minority view is objectively without merit”).
81
55
constitutional interpretation. Interpretive philosophy determines
flagrancy of error, and flagrancy of error determines susceptibility to
overruling. If precedents are only as strong as the interpretive
commitments of five Supreme Court justices, the continuity and
impersonality of constitutional law is impaired.
Rather than treating all precedents that arise from rival
interpretive schools as egregiously wrong, the justices might show
interpretive empathy by placing themselves within the decisional
mindset a prior decision reflects. An originalist justice would not treat
flawed decisions as manifestly erroneous simply because they were
reasoned on pragmatic grounds. Instead, she would evaluate those
decisions against a backdrop assumption of pragmatism’s validity. Nor
would a pragmatic justice treat all originalist precedents as manifestly
erroneous. Instead, she would assess the flagrancy of error from the
standpoint of an originalist. This practice would preserve a distinction
between the adoption of a given interpretive philosophy and the
declaration of manifest error.
But interpretive empathy presents difficulties of its own. Assuming
that the justices have the time and capacity to deploy divergent
methodologies from case to case, interpretive empathy requires them
to embrace, at least hypothetically, an interpretive philosophy they
might view as imprudent or illegitimate. An originalist justice is
required to channel a competing methodology that is incompatible with
the tenets of originalism. The same is true of the pragmatist or
common law constitutionalist, who must imagine himself as a
momentary originalist regardless of any doubts he may harbor about
the originalist enterprise.
Additional problems arise with respect to precedents that do not fit
neatly into one interpretive box. Supreme Court decisions often
describe different types of arguments as pointing toward the same
result. 83 That practice can render it impracticable for a later justice to
adopt the methodological mindset of the Court that issued the
precedent: There may be no way to determine what, exactly, that
mindset was.
See Richard H. Fallon, Jr., A Constructivist Coherence Theory of Constitutional
Interpretation, 100 HARV. L. REV. 1189, 1192–93 (1987).
83
56
In light of the difficulties associated with interpretive fidelity and
interpretive empathy, the flagrancy of a precedent’s error fits uneasily
with the doctrine of stare decisis in a pluralistic legal culture. The
egregiousness of a precedent’s error coheres with the doctrine of stare
decisis only in a world of interpretive agreement. Second-best stare
decisis must look elsewhere for its content. It is possible, if unlikely,
that a Supreme Court opinion may be not simply wrong but
illegitimate and unlawful, for instance because it was written by a
justice who expressly ignored the relevant enactments and ruled based
on personal affinity or a flip of the coin. Such decisions, of course, are
entitled to no deference whatsoever. But beyond that category of
extreme cases, second-best stare decisis focuses on factors apart from
theory-dependent conclusions about the soundness of a decision’s
reasoning.
Substantive Effects
The importance of deciding a case correctly naturally is informed by
the ramifications of deciding it incorrectly. 84 Does it follow that
precedents with undesirable consequences should have a limited shelf
life?
I submit that the answer is no. The assessment of a precedent’s
harmfulness depends on conclusions about legal relevance, which
themselves depend on theories of constitutional interpretation.
Substantive assessments—and the methodological and normative
commitments they reflect—are therefore unsuitable for a world of
interpretive disagreement. Second-best stare decisis responds by
excluding substantive effects from the overruling calculus in the
ordinary course. In selecting the best interpretation of a disputed
constitutional provision, the justices will continue to disagree about
the types of benefits and harms that are legally relevant. Second-best
stare decisis has no objection to that phenomenon. It simply prevents
such disagreements from reemerging when the justices turn to the
distinct question whether a flawed precedent should be retained or
jettisoned.
Cf. Henry Paul Monaghan, Stare Decisis and Constitutional Adjudication, 88
COLUM. L. REV. 723, 760 (1988) (“The most difficult question in identifying the
harmful effects due to a wrongly decided controlling precedent is whether this
criterion can be rendered sufficiently principled so that it is not simply a euphemism
describing decisions that a Court majority very much dislikes.”).
84
57
Treating substantive effects as inapposite reflects a compromise
that cuts across methodological and ideological lines. 85 Justices who
would otherwise be inclined to view a precedent’s injustice or inefficacy
as a reason for overruling must set those factors aside. Likewise,
justices who emphasize values such as popular sovereignty will need to
mediate their view that precedents are in greater need of overruling
when they impede democratic self-government; the primacy of popular
sovereignty is not a value that extends across interpretive
methodologies. These are just a few illustrations of a general point:
Determining which types of effects are legally relevant is theorydependent by definition, making it unsuitable for a doctrine of stare
decisis designed to bridge interpretive disputes. Conclusions about the
salient effects of flawed decisions can serve as the basis for declaring a
precedent to be erroneous (or for resolving a case of first impression),
but they are generally excluded from the inquiry into a precedent’s
retention. Rather than drawing on the substantive value of
interpreting the Constitution correctly, second-best stare decisis
revolves around factors whose application can be cordoned off from
disputes over interpretive theory.
Disregard of substantive effects characterizes the ordinary course of
second-best stare decisis. But the ordinary course is not the only course,
and there is a category of exceptional situations in which a precedent’s
substantive effects may play a legitimate role in the second-best
analysis. Cases may occasionally arise in which a justice perceives an
overwhelming justification for renouncing a precedent due to its
substantive effects. The common law constitutionalist might view a
precedent as not merely unfair, but profoundly immoral. The
pragmatist might view a precedent as not merely ill-advised, but
disastrous in its practical implications. The popular sovereignty
originalist might view a precedent as not merely anti-democratic, but
an intolerable affront to the authority of the people. And so on.
Despite its general exclusion of substantive effects, second-best
stare decisis acknowledges the legitimacy of substantive considerations
John McGinnis and Michael Rappaport strike a similar tone in their
consequentialist defense of originalism: “[A] constitution in a pluralistic society
should rest to some extent on a compromise among people’s views of the good in that
society.” JOHN O. MCGINNIS & MICHAEL B. RAPPAPORT, ORIGINALISM AND THE GOOD
CONSTITUTION 5 (2013).
85
58
in these exceptional cases. Fidelity to precedent is not absolute. It is a
presumption, subject to override when certain criteria are met. In most
cases, those criteria should exclude a precedent’s substantive effects in
order to prevent stare decisis from being mired in debates over
interpretive philosophy. Yet second-best stare decisis can recognize
exceptional cases in which substantive effects are relevant without
jeopardizing the larger project of accommodating the treatment of
precedent to a pluralistic world.
The key is ensuring that attention to substantive effects remains
the exception rather than the rule. Invoking substantive
considerations to justify an overruling triggers a corresponding
obligation to imagine what the consequences would be if one’s judicial
colleagues were to behave in the same way with respect to precedents
they view as severely problematic. This type of aggregated analysis
sharply limits the category of cases in which substantive
considerations are relevant.
For example, even under the second-best approach, a justice could
leave open the possibility that considerations of morality can be
relevant to the durability of precedent. But she would reserve those
considerations for extreme cases rather than allowing them to
permeate the doctrine of stare decisis in every constitutional dispute. A
different justice might determine that precedents posing a substantial
threat to popular sovereignty—perhaps by inhibiting the free elections
necessary for meaningful self-government—should be overturned
based on their substantive effects. Still other justices might conclude
the effects of constitutional mistakes are most severe when the Court
has overstepped its proper role by intervening in a divisive political or
social debate. Whichever of these (or other) conceptual lenses a
particular justice employs, second-best stare decisis forecloses the
broader conclusion that all erroneous precedents are subject to
reconsideration based on their substantive effects. While that latter
approach may be sensible in a world of interpretive agreement, it fails
under conditions of interpretive pluralism.
Acknowledging the relevance of substantive effects in extraordinary
cases also makes second-best stare decisis more plausible in practical
terms. Sitting and future justices might well balk upon being asked to
uphold what they view as the worst of the worst precedents. The
substantive-effects exception responds by acting as a safety valve. The
59
exception provides space for correcting the Court’s most harmful
(however that concept is defined) mistakes without creating pressure
to distort the ordinary tools of stare decisis analysis. So long as
substance-based overrulings are limited to rare and exceptional
situations, they do not threaten the enterprise of second-best stare
decisis. To the contrary, the doctrine operates as it should: by creating
a strong presumption that yields when countervailing considerations
are truly compelling.
Second-Best Stare Decisis Beyond the Core Factors
Thus far I have discussed the implications of second-best analysis
for the rules of stare decisis as described in Planned Parenthood of
Southeastern Pennsylvania v. Casey, which represents the Supreme
Court’s most notable prominent of the doctrine. I have also applied the
second-best approach to two other factors—the soundness of a
precedent’s reasoning and the harmfulness of its substantive effects—
that often emerge in disputes over the durability of precedent. Taken
together, these considerations are properly understood as the core of
the Court’s modern approach to stare decisis. Still, other
considerations also crop up from time to time in the Court’s discussions
of precedent. I will briefly discuss three of those factors as further
examples of how second-best analysis applies to the doctrine of stare
decisis.
Age. Older precedents sometimes receive heightened deference in
light of their vintage. Their staying power over the years may have
proved them to be sound. Older precedents have also had more time to
generate reliance by stakeholders. It follows, the argument goes, that
older precedents should receive more respect than newer ones—the
latter of which can be corrected “before state and federal laws and
practices have been adjusted to embody” them. 86
The role of a precedent’s age is complicated by Supreme Court
opinions privileging newer opinions over older ones. This often occurs
in connection with the inquiry into jurisprudential coherence, which I
discussed above. When the Court depicts prior decisions as having
South Carolina v. Gathers, 490 U.S. 805, 824 (1989) (Scalia, J., dissenting);
Mitchell v. Helms, 530 U.S. 793, 835–36 (2000) (plurality op.); see also, e.g., Montejo,
556 U.S. at 792; STEPHEN BREYER, MAKING OUR DEMOCRACY WORK: A JUDGE’S VIEW
152 (2010).
86
60
been undermined by more recent ones, the effect is to elevate the new
over the old. To similar effect is Justice Ginsburg’s statement about
being faithful to “later, more enlightened decisions” and while
jettisoning past decisions that have become “outworn.” 87 The principle
that newer opinions can “undermin[e]” the “doctrinal underpinnings”
of older ones means that in some cases, the former take priority over
the latter. 88
The overruling of recent decisions can also be troubling from the
standpoint of judicial impersonality. Recall Justice Marshall’s dissent
in Payne v. Tennessee (1991)—a case dealing with the use of victimimpact evidence—in which he criticized the majority for overruling
recent precedents even though “[n]either the law nor the facts,” but
“[o]nly the personnel of this Court,” had changed. 89 Abrupt overrulings
following changes the Court’s composition can blur the line between
the meaning of the Constitution and the identity of the individuals
who occupy the bench. It reinforces both the perception and reality
that constitutional change occurs primarily through the judicial
appointment process rather than the Article V amendment process. As
Justice Stevens once observed, “Citizens must have confidence that the
rules on which they rely in ordering their affairs . . . are rules of law
and not merely the opinions of a small group of men who temporarily
occupy high office.” 90
The problem is not that a precedent’s vintage is unsuitable for
consideration under a second-best approach. To the contrary, vintage is
the type of objective, independent factor that lends itself to application
by justices across the methodological spectrum. The problem is the
Court has not consistently treated earlier precedents as more
venerable then later ones, nor has it offered a convincing account of
why overrulings are more problematic when precedents are old. The
better approach is to focus directly on issues such a precedent’s
reliance implications without filtering the analysis through the foggy
lens of a precedent’s age.
John R. Sand & Gravel Co. v. United States, 552 U.S. 130, 144 (2008) (Ginsburg,
J., dissenting); Lawrence, 539 U.S. at 573–76.
88 See, e.g., Dickerson, 530 U.S. at 443.
89 Payne, 501 U.S. at 844 (Marshall, J., dissenting).
90 Florida Department of Health and Rehabilitative Services v. Florida Nursing
Home Association, 450 U.S. 147, 154 (1981) (Stevens, J., concurring).
87
61
Dissent and Criticism. A case’s divisiveness can affect its staying
power. Deference is reduced when a precedent was “decided by the
narrowest of margins, over spirited dissents challenging [its] basic
underpinnings.” 91 Along similar lines, the justices sometimes ask
whether a precedent has drawn criticism from the bench and scholarly
community, which can affect the extent to which the precedent has
generated reliance expectations among stakeholders. 92 The inverse
occurred in Casey, where the lead opinion described the criticism faced
by Roe as a reason for standing by that decision. 93 But the Casey Court
made clear that its rationale reflected the exceptional degree of public
interest in Roe. In the ordinary course, it is more common to find
criticism as cutting against deference rather than in its favor.
In defending this practice, we might posit that divided opinions are
more likely to occur in difficult and divisive cases. Such cases, in turn,
are relatively likely to result in unsound decisions. There is also a
reliance angle: All else equal, perhaps stakeholders are (or should be)
less likely to rely on decisions that are hard-fought and closely divided.
By definition, those decisions are the fewest votes away from
overruling. And if several justices felt strongly enough about an issue
to dissent in dramatic fashion, they may be more likely to persist in
their disagreement as they try to win a majority through persuasion or
attrition.
There is another side to the story. It is possible that the majority
opinion in a divisive case is particularly likely to be soundly reasoned,
precisely because the competing arguments received such extensive
attention. The Supreme Court took this position in Patterson v.
McLean Credit Union (1989), where part of its justification for refusing
to overrule a statutory precedent owed to the fact that “[t]he
arguments about whether [the precedent] was decided correctly in
light of the language and history of the statute were examined and
discussed with great care in our decision.” The Court explained that
“[i]t was recognized at the time” of the relevant decision “that a strong
case could be made for the” opposing view.” Yet “that view did not
prevail,” and the Court saw no reason to reopen the debate. 94 Justice
Marshall made a similar point in his Payne dissent, which criticized
Payne, 501 U.S. at 828–29 (majority op.).
See, e.g., Citizens United, 558 U.S. at 380 (Roberts, C.J., concurring).
93 See Casey, 505 U.S. at 867.
94 Patterson v. McLean Credit Union, 491 U.S. 164, 171 (1989).
91
92
62
the majority for departing from precedent based on arguments the
Court had recently rejected. 95 Understood in this way, contentious
decisions represent the antithesis of the lightly considered statements
the Court has described as unworthy of deference due to the superficial
deliberation they appear to have received. 96
On balance, neither the presence of a spirited dissent nor the
persistence of criticism is a dependable proxy for a decision’s likelihood
of error on the merits. Moreover, if dissents and criticism are meant as
proxies for a decision’s effect on reliance expectations, it is better to
evaluate reliance directly. That leaves little reason for infusing dissent
and criticism with independent force in the stare decisis analysis.
Nature of Decisional Rule. The type of rule an opinion announced
affects the deference it receives. I noted in Chapter 1 that the Supreme
Court continues to maintain a distinction between its statutory and
constitutional precedents, with the former receiving a higher degree of
deference. The Court also gives heightened deference to decisions in
fields such as property and contract law, on the rationale that those
decisions often have the greatest impact on reliance expectations. By
contrast, rules of evidence and procedure receive diminished deference.
Those types of rules do “not affect the way in which parties order their
affairs,” and their revision does “not upset settled expectations on
anyone’s part.” 97
Inquiry into the nature of a decisional rule is compatible with the
principles of second-best stare decisis. Whatever their methodological
and normative inclinations, the justices will often be able to agree
about whether a rule is substantive or procedural. The more difficult
question is whether the nature of decisional rule reveals anything
meaningful about the deference owed to a precedent.
It is possible that the nature of a decisional rule correlates with
reliance expectations and, thus, with the disruption an overruling
would entail. But that is not always the case. Take Hohn v. United
States (1998). There, the Supreme Court considered whether it had
jurisdiction to review denials of certifications of appealability, which
petitioners in federal habeas corpus proceedings need in order to
See Payne, 501 U.S. at 846 (Marshall, J., dissenting).
See, e.g., McCutcheon v. FEC, 134 S. Ct. 1434, 1446 (2014) (plurality op.).
97 Pearson v. Callahan, 555 U.S. 223, 233 (2009).
95
96
63
appeal from an adverse district court decision. The Court found itself
to possess the requisite jurisdiction, but it acknowledged that its
conclusion conflicted with a precedent issued some fifty years prior.
Explaining its decision to overrule the precedent, the Court cited its
practice of giving reduced deference to procedural rules. In dissent,
Justice Scalia agreed that “procedural rules do not ordinarily engender
detrimental reliance.” He disagreed with the majority because the case
at hand was not ordinary. Even if the relevant precedent had not
affected primary conduct by private citizens, Congress had presumably
paid attention to it. Irrespective of its procedural nature, the precedent
warranted preservation in light of the “reliance of Congress upon an
unrepudiated decision central to the procedural scheme it was
creating.” 98
Whatever the proper result in Hohn, the colloquy between majority
and dissent shows why the nature of a precedent’s decisional rule is
not always an accurate proxy for reliance and disruption. Sometimes
procedural and evidentiary do not engender reliance, and sometimes
they do. In light of that variability, the better course is the direct
analysis of reliance interests. Again, this is not because the nature of a
decisional rule is out of bounds for second-best stare decisis. The fact
that a particular feature of a case has independent content apart from
debates over interpretive methodology is necessary for its inclusion in
the second-best analysis, but it is not sufficient. The feature must also
tell us something meaningful about when precedents should be
retained and when they should be overruled.
Collective Courts and Individual Judges
A theory of adjudication that would be desirable if endorsed by a
majority of Supreme Court justices might lose its resonance if endorsed
by only a single justice. The ultimate question is not how the Court as
an entity should proceed. It is what “decision procedure should a ninemember body employ to reach the best decisions they can over an array
of cases highlighting experiential and political differences among the
nine decisionmakers.” 99 The point is relevant to second-best stare
Hohn v. United States, 524 U.S. 236, 259–61 (1998) (Scalia, J., dissenting).
Frederick Schauer, Statutory Construction and the Coordinating Function of Plain
Meaning, 1990 SUP. CT. REV. 231, 255; see also ADRIAN VERMEULE, THE SYSTEM OF
THE CONSTITUTION 137 (2011) (“Even if all judges should adopt a given theory, it does
not follow that any one should, because others may not.”).
98
99
64
decisis, for it raises the question whether an individual justice should
compromise her vision of optimal legal interpretation absent
assurances that her colleagues (and successors) will follow suit.
There is no guarantee a Supreme Court justice’s application of
second-best stare decisis will convince her colleagues to take the same
path. Nor is there any guarantee that the adoption of second-best stare
decisis in a majority opinion will entrench that approach in perpetuity.
Nevertheless, justices who apply second-best stare decisis can help to
establish its prevalence going forward. And even if a justice finds
herself paying greater attention to precedent than her colleagues do,
she can still make a valuable contribution in her own right. Adopting
the second-best approach is about tempering one’s interpretive
philosophy for the sake of behaving like a member of an impersonal
and enduring institution. To be sure, the benefits of stare decisis are
greatest when a majority of justices endorse it. But a single justice who
casts her lot with stare decisis can promote the ideals of continuity and
impersonality. The act of compromising one’s interpretive predilections
underscores the separation between judge and law, as well as the ideal
that legal rules are general norms to which courts commit themselves
across the span of time. A vote for second-best stare decisis represents
a contribution to the rule of law irrespective of the conduct of one’s
judicial peers. 100 For a single justice, a vote to uphold precedent in a
particular case is a means of strengthening a doctrine that provides
value to the constitutional system more broadly.
Analyzing a justice’s opportunity costs—in other words, what she
gives up by supporting stare decisis—leads to the same conclusion. To
illustrate, consider three scenarios that might confront a precedentfriendly Supreme Court justice. In the first scenario, most of her
colleagues have decided to overrule an applicable precedent. A few
justices have resisted that conclusion, but not on stare decisis grounds;
they simply think the majority is mistaken on the merits. Our justice
believes the majority has the better of the argument, but she would
preserve the applicable precedent for reasons of stare decisis. In this
situation, our justice could choose to abandon stare decisis by adding
Cf. ADRIAN VERMEULE, JUDGING UNDER UNCERTAINTY 146 (2006) (noting that “[s]o
long as the relevant theory does not require or assume a critical mass or threshold of
judicial coordination—so long as individual judges may make a strictly divisible or
marginal contribution to the aims specified by the theory—then the infeasibility of
sustained judicial coordination poses no problem”).
100
65
her voice to the chorus of justices in the majority. Yet that approach is
unlikely to affect the impact of the Court’s opinion, which already
enjoys majority support. Our justice thus gives up little by writing a
dissent on grounds of stare decisis.
In making this choice, the justice does not alter the outcome of the
case at hand. But she does promote the principle that justices are
willing to stand by precedents despite their personal misgivings. If our
justice believes it advances the rule of law when judges subordinate—
and are seen as subordinating—their individual views for the sake of
keeping faith with their court’s institutional history, she makes a
contribution by deferring to precedent even if her colleagues do not
follow her lead. Likewise, if our justice believes deference to precedent
is a useful second-best principle for facilitating coordinated action, her
dissenting opinion may enhance the viability of stare decisis in future
cases by keeping it on the Court’s agenda. In all events, her choice to
dissent carries little cost given that her vote did not affect the outcome
of the case. 101
In the second scenario, most of the Court’s members believe the
precedent under consideration is correct. Our justice disagrees with
that conclusion on the merits, but she thinks stare decisis requires
upholding the precedent. Our justice could eschew considerations of
stare decisis and dissent, but her choice would not change the outcome
of the case. On the other hand, our justice could concur in the
majority’s result, issuing a separate opinion to explain that her vote is
based on stare decisis. Again, this course of action contributes to the
vitality of stare decisis going forward while imposing little cost.
Finally, imagine a scenario in which our justice’s colleagues are
evenly split. She is the “swing” vote, with the power to control the
outcome of the case. There is an applicable precedent on the books, but
none of her colleagues give it much import. Instead, they reach their
decisions purely on the merits. Now our justice has the opportunity to
The Supreme Court occasionally states that closely divided decisions are entitled
to less deference going forward. On that logic, an opinion supported by six votes may
receive marginally greater deference in future cases than an opinion supported by
five votes. I have criticized this practice as unjustified. Even if the practice persists,
it is difficult to foresee a situation in which five justices are prepared to overrule a
decision but refrain from doing so because it received six votes rather than five, or
seven rather than six, and so on.
101
66
infuse the doctrine of stare decisis with considerable salience. She can
explain that regardless of her views on the merits, she believes proper
application of stare decisis requires abiding by the applicable
precedent even if it is incorrect. In this way, our justice demonstrates
that stare decisis affects not just rationales but results. She succeeds
in bringing about the proper outcome (from the perspective of stare
decisis) in the case at hand. In doing so, she also strengthens the
general practice of precedent-following on the Court. Whatever
happens in the years ahead, she will have demonstrated that stare
decisis transcends interpretive and normative commitments.
These scenarios do not exhaust the possibilities. Some cases might
lead to a wider array of positions among smaller factions of justices.
But in the common scenarios I have described, the best course for our
justice will be to apply second-best stare decisis regardless of what her
colleagues choose to do. 102 Our justice can treat precedent with
presumptive deference without worrying about her colleagues’ votes.
And if her colleagues ultimately adopt the same mindset as her, we
have the framework for a meaningful doctrine of stare decisis in
Supreme Court decisionmaking.
A general practice of precedent-following might also arise from the
justices’ belief that if they wish for their preferred precedents to stand
the test of time, they must be willing to abide by some decisions with
which they disagree. Adrian Vermeule explains how judges who are
involved in repeated interactions might be inclined to defer to
precedent to encourage reciprocation from their colleagues. 103 Michael
Gerhardt makes a related point in depicting the justices as recognizing
that by showing disdain for disfavored precedents, they put their
preferred precedents on shakier ground. 104
To the extent justices think in such strategic terms, the likelihood
increases of creating a meaningful doctrine of stare decisis as a
mechanism for protecting the good (from the perspective of an
individual justice) at the cost of enduring some bad. But even if the
justices disregard these strategic considerations—either because they
I hasten to reiterate that applying second-best stare decisis does not always mean
abiding by precedent; there are situations in which precedent should yield.
103 See VERMEULE, THE SYSTEM OF THE CONSTITUTION, supra note __, at 142.
104 See MICHAEL J. GERHARDT, THE POWER OF PRECEDENT 79 (2008)
102
67
find the prospect of retaliation to be unsuitable to the Court’s work, or
because they regard the Court’s future as sufficiently unpredictable
that credible threats of retaliation are undermined—there remains the
argument I developed above: A vote in favor of stare decisis, even by a
single justice, strengthens the doctrine going forward, enhancing its
ability to generate benefits such as impersonality, stability, and the
facilitation of coordinated action.
While tomorrow’s justices always possess the power to depart from
particular precedents or even to abandon the doctrine of stare decisis
entirely, today’s justices can make such developments less likely. A
justice who operates within a practice of deferring to precedent faces
an “argumentative burden” in justifying her choice to overrule. 105 She
must explain why she perceives a suitable reason for departing from
precedent—or why she thinks no such reason is required despite her
predecessors’ claims to the contrary. If she doubts her ability to make
these arguments in a persuasive manner, she may be inclined to defer
to a precedent even if she questions its soundness. That result would
only come to pass in a system within which deference to precedent is
the customary practice. By respecting precedent today, a justice makes
it more likely that precedent will command respect tomorrow.
There is also the question whether precedent-friendly justices
should be wary of “ratchet” effects that threaten to chip away at their
preferred vision of constitutional law. 106 The concern is that if some
justices are more inclined than others to defer to precedent, and if that
distinction tracks differences in interpretive approach, over time the
law may move toward the view preferred by those who eschew stare
decisis. Foreseeing this possibility, justices who would otherwise defer
to precedent might consider voting to overrule most or all decisions
that reflect an interpretive methodology different from their own.
Such a choice would, I think, be unwise. Though it may be possible
to make educated guesses about how one justice’s treatment of
precedent is likely to be received—and responded to—by her colleagues
in the immediate future, the long-term effects on the Court’s caselaw
are extremely difficult to predict. To begin, it is not necessarily correct
Cf. Frederick Schauer, Precedent, 39 Stan. L. Rev. 571, 580 (1987).
Lawrence B. Solum, The Supreme Court in Bondage: Constitutional Stare Decisis,
Legal Formalism, and the Future of Unenumerated Rights, 9 U. PA. J. CONST. L. 155,
193 (2006).
105
106
68
to view judicial philosophies as static. A commitment to precedent by
proponents of one philosophy might have dynamic effects in
encouraging proponents of other philosophies to become more receptive
to decisions with which they disagree—leading to the establishment of
a meaningful doctrine of stare decisis across interpretive schools.
There is also the possibility that stare decisis might create shortterm costs but long-term gains for its adherents. An originalist justice
who upholds a nonoriginalist decision might make a marginal
contribution to the salience of nonoriginalist constitutional rules. But
perhaps the originalist justice who stands by precedent will contribute
to a perception of originalism as linked with constancy and
impersonality, eventually increasing the prominence of the originalist
school and leading to the appointment of more originalist justices.
Cases of first impression would henceforth be decided according to
originalist principles, and originalist precedents would be insulated
from overruling.
Nor is it clear that committing oneself to stare decisis will privilege
one interpretive methodology at the expense of others. Even if future
justices are inclined to overrule precedents that conflict with their
preferred constitutional theories, there is no reason to think such
justices will disproportionately come from a particular methodological
school. It is plausible that the Court could vacillate between periods in
which originalist justices regularly overruled nonoriginalist precedents,
and periods in which living constitutionalist justices regularly
overruled originalist precedents. The aggregate effect would be a
certain amount of overruling, but it would not be a ratchet that
steadily entrenches one theory while undermining others. The ratchet
concern would arise only if frequent overruling were correlated with
the prevalence of a particular interpretive philosophy. That need not
be the case. Indeed, as I have argued, there is ample room in theories
as diverse as originalism and living constitutionalism for meaningful
deference to precedents—even precedents that are incorrect when
viewed through a given methodological lens.
It remains true that by deferring to a flawed precedent, today’s
justice might act in a way that some of her colleagues would not. She
might also sacrifice the power to right a constitutional wrong. The
extent of the sacrifice is reduced by the fact that second-best stare
decisis preserves the justice’s ability to rectify the worst of the worst
69
constitutional errors even at the cost of continuity. 107 But the more
important point is that the power to rectify perceived constitutional
mistakes is fleeting in a world without stare decisis. Such power lasts
only until the Court’s balance of power shifts, at which time a new
majority is in position to dispense with contrary precedents. Fidelity to
precedent, by contrast, promotes coordinated action among differentlyminded jurists regardless of what happens in the years ahead. By
helping to establish second-best stare decisis as an influential doctrine,
even a single justice can make it harder for future justices to depart
from precedent—and, as a result, less likely that they will try.
Concerns about judicial disregard for precedent do not only face
forward. They also encourage today’s Court to look backward for times
at which prior justices have given precedent less than its due. Consider,
for example, those who believe the Warren Court was unfaithful to the
Court’s prior teachings on criminal procedure 108 or the Rehnquist
Court contravened established caselaw on federalism. A Supreme
Court justice who possesses such a belief faces a choice. She may abide
by the existing precedents, unfaithful as they may have been to older
cases. Or she may engage in her own overruling with the goal of
restoring the legal rules that were improperly supplanted.
There is support in the Supreme Court’s caselaw for the conclusion
that an opinion borne of inadequate respect for its ancestors should
expect the same irreverent treatment from its heirs. Hence the Court’s
willingness to overrule opinions that “deviated sharply from our
established . . . jurisprudence” 109 or that represented a jurisprudential
“aberration.” 110 Yet second-best stare decisis suggests the better course
is to stand by existing precedents regardless of what came before. The
crucial moment is the present one. The choice before today’s justice is
whether to vote in favor of reaffirming or overruling. That choice
should occur within the framework of second-best stare decisis, which
avoids theory-laden determinations of whether a precedent was
faithful to the cases that came before it.
Second-best stare decisis is not aligned with any interpretive
methodology. It will counsel against overruling some precedents that
See Chapter 6.
VERMEULE, THE SYSTEM OF THE CONSTITUTION, supra note __, at 144.
109 Seminole Tribe of Florida v. Florida, 517 U.S. 44, 64 (1996).
110 Citizens United v. FEC, 558 U.S. 310, 355 (2010).
107
108
70
are disfavored by originalists, but it will do the same with respect to
precedents that are disfavored by living constitutionalists. A
commitment to stare decisis is not a commitment to originalism, living
constitutionalism, or any other theory. It is a commitment to the
abiding continuity of constitutional law even as individual justices
come and go.
71